{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 5.827814569536423,
  "eval_steps": 50,
  "global_step": 1100,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "classification_loss": 0.7052,
      "epoch": 0.005298013245033113,
      "grad_norm": 16.536712646484375,
      "learning_rate": 3.5398230088495575e-07,
      "lm_loss": 2.0475,
      "loss": 2.7527,
      "step": 1
    },
    {
      "classification_loss": 0.8302,
      "epoch": 0.010596026490066225,
      "grad_norm": 19.213396072387695,
      "learning_rate": 7.079646017699115e-07,
      "lm_loss": 1.9993,
      "loss": 2.8294,
      "step": 2
    },
    {
      "classification_loss": 0.7851,
      "epoch": 0.015894039735099338,
      "grad_norm": 21.516714096069336,
      "learning_rate": 1.0619469026548673e-06,
      "lm_loss": 1.9864,
      "loss": 2.7715,
      "step": 3
    },
    {
      "classification_loss": 0.7105,
      "epoch": 0.02119205298013245,
      "grad_norm": 18.66866111755371,
      "learning_rate": 1.415929203539823e-06,
      "lm_loss": 2.0032,
      "loss": 2.7137,
      "step": 4
    },
    {
      "classification_loss": 0.7335,
      "epoch": 0.026490066225165563,
      "grad_norm": 18.92946434020996,
      "learning_rate": 1.769911504424779e-06,
      "lm_loss": 2.1268,
      "loss": 2.8603,
      "step": 5
    },
    {
      "classification_loss": 0.7336,
      "epoch": 0.031788079470198675,
      "grad_norm": 19.057477951049805,
      "learning_rate": 2.1238938053097345e-06,
      "lm_loss": 1.9969,
      "loss": 2.7305,
      "step": 6
    },
    {
      "classification_loss": 0.8332,
      "epoch": 0.03708609271523179,
      "grad_norm": 22.374317169189453,
      "learning_rate": 2.4778761061946905e-06,
      "lm_loss": 1.8906,
      "loss": 2.7238,
      "step": 7
    },
    {
      "classification_loss": 0.6982,
      "epoch": 0.0423841059602649,
      "grad_norm": 15.95816421508789,
      "learning_rate": 2.831858407079646e-06,
      "lm_loss": 1.9744,
      "loss": 2.6726,
      "step": 8
    },
    {
      "classification_loss": 0.8174,
      "epoch": 0.04768211920529802,
      "grad_norm": 16.178144454956055,
      "learning_rate": 3.185840707964602e-06,
      "lm_loss": 1.8469,
      "loss": 2.6643,
      "step": 9
    },
    {
      "classification_loss": 0.7296,
      "epoch": 0.052980132450331126,
      "grad_norm": 20.708654403686523,
      "learning_rate": 3.539823008849558e-06,
      "lm_loss": 1.765,
      "loss": 2.4947,
      "step": 10
    },
    {
      "classification_loss": 0.7078,
      "epoch": 0.05827814569536424,
      "grad_norm": 13.415894508361816,
      "learning_rate": 3.8938053097345135e-06,
      "lm_loss": 1.4819,
      "loss": 2.1896,
      "step": 11
    },
    {
      "classification_loss": 0.5996,
      "epoch": 0.06357615894039735,
      "grad_norm": 15.516067504882812,
      "learning_rate": 4.247787610619469e-06,
      "lm_loss": 1.444,
      "loss": 2.0437,
      "step": 12
    },
    {
      "classification_loss": 0.7094,
      "epoch": 0.06887417218543046,
      "grad_norm": 12.468376159667969,
      "learning_rate": 4.6017699115044254e-06,
      "lm_loss": 1.1518,
      "loss": 1.8612,
      "step": 13
    },
    {
      "classification_loss": 0.7456,
      "epoch": 0.07417218543046358,
      "grad_norm": 16.213367462158203,
      "learning_rate": 4.955752212389381e-06,
      "lm_loss": 1.0547,
      "loss": 1.8003,
      "step": 14
    },
    {
      "classification_loss": 0.7801,
      "epoch": 0.07947019867549669,
      "grad_norm": 12.886180877685547,
      "learning_rate": 5.309734513274337e-06,
      "lm_loss": 0.9856,
      "loss": 1.7656,
      "step": 15
    },
    {
      "classification_loss": 0.7944,
      "epoch": 0.0847682119205298,
      "grad_norm": 13.143818855285645,
      "learning_rate": 5.663716814159292e-06,
      "lm_loss": 0.8596,
      "loss": 1.6541,
      "step": 16
    },
    {
      "classification_loss": 0.7557,
      "epoch": 0.09006622516556291,
      "grad_norm": 11.865086555480957,
      "learning_rate": 6.0176991150442484e-06,
      "lm_loss": 0.7187,
      "loss": 1.4744,
      "step": 17
    },
    {
      "classification_loss": 0.7017,
      "epoch": 0.09536423841059603,
      "grad_norm": 13.191651344299316,
      "learning_rate": 6.371681415929204e-06,
      "lm_loss": 0.6845,
      "loss": 1.3862,
      "step": 18
    },
    {
      "classification_loss": 0.718,
      "epoch": 0.10066225165562914,
      "grad_norm": 11.245887756347656,
      "learning_rate": 6.72566371681416e-06,
      "lm_loss": 0.5743,
      "loss": 1.2923,
      "step": 19
    },
    {
      "classification_loss": 0.7031,
      "epoch": 0.10596026490066225,
      "grad_norm": 10.283029556274414,
      "learning_rate": 7.079646017699116e-06,
      "lm_loss": 0.4685,
      "loss": 1.1716,
      "step": 20
    },
    {
      "classification_loss": 0.6872,
      "epoch": 0.11125827814569536,
      "grad_norm": 11.32901382446289,
      "learning_rate": 7.4336283185840714e-06,
      "lm_loss": 0.3906,
      "loss": 1.0778,
      "step": 21
    },
    {
      "classification_loss": 0.7607,
      "epoch": 0.11655629139072848,
      "grad_norm": 10.964323043823242,
      "learning_rate": 7.787610619469027e-06,
      "lm_loss": 0.3585,
      "loss": 1.1192,
      "step": 22
    },
    {
      "classification_loss": 0.7029,
      "epoch": 0.12185430463576159,
      "grad_norm": 17.21455192565918,
      "learning_rate": 8.141592920353984e-06,
      "lm_loss": 0.3023,
      "loss": 1.0053,
      "step": 23
    },
    {
      "classification_loss": 0.7164,
      "epoch": 0.1271523178807947,
      "grad_norm": 12.168242454528809,
      "learning_rate": 8.495575221238938e-06,
      "lm_loss": 0.2346,
      "loss": 0.9509,
      "step": 24
    },
    {
      "classification_loss": 0.6726,
      "epoch": 0.13245033112582782,
      "grad_norm": 13.1473388671875,
      "learning_rate": 8.849557522123895e-06,
      "lm_loss": 0.2101,
      "loss": 0.8827,
      "step": 25
    },
    {
      "classification_loss": 0.7433,
      "epoch": 0.13774834437086092,
      "grad_norm": 11.16681957244873,
      "learning_rate": 9.203539823008851e-06,
      "lm_loss": 0.1793,
      "loss": 0.9225,
      "step": 26
    },
    {
      "classification_loss": 0.6998,
      "epoch": 0.14304635761589404,
      "grad_norm": 10.06042194366455,
      "learning_rate": 9.557522123893806e-06,
      "lm_loss": 0.1519,
      "loss": 0.8518,
      "step": 27
    },
    {
      "classification_loss": 0.6573,
      "epoch": 0.14834437086092717,
      "grad_norm": 17.411531448364258,
      "learning_rate": 9.911504424778762e-06,
      "lm_loss": 0.0985,
      "loss": 0.7558,
      "step": 28
    },
    {
      "classification_loss": 0.7481,
      "epoch": 0.15364238410596026,
      "grad_norm": 12.332111358642578,
      "learning_rate": 1.0265486725663717e-05,
      "lm_loss": 0.1002,
      "loss": 0.8483,
      "step": 29
    },
    {
      "classification_loss": 0.6751,
      "epoch": 0.15894039735099338,
      "grad_norm": 13.662885665893555,
      "learning_rate": 1.0619469026548675e-05,
      "lm_loss": 0.0833,
      "loss": 0.7584,
      "step": 30
    },
    {
      "classification_loss": 0.6691,
      "epoch": 0.16423841059602648,
      "grad_norm": 8.711078643798828,
      "learning_rate": 1.0973451327433629e-05,
      "lm_loss": 0.087,
      "loss": 0.7561,
      "step": 31
    },
    {
      "classification_loss": 0.731,
      "epoch": 0.1695364238410596,
      "grad_norm": 9.07862377166748,
      "learning_rate": 1.1327433628318584e-05,
      "lm_loss": 0.0958,
      "loss": 0.8267,
      "step": 32
    },
    {
      "classification_loss": 0.6431,
      "epoch": 0.17483443708609273,
      "grad_norm": 7.65638542175293,
      "learning_rate": 1.1681415929203541e-05,
      "lm_loss": 0.0758,
      "loss": 0.7189,
      "step": 33
    },
    {
      "classification_loss": 0.6956,
      "epoch": 0.18013245033112582,
      "grad_norm": 7.838529586791992,
      "learning_rate": 1.2035398230088497e-05,
      "lm_loss": 0.0887,
      "loss": 0.7843,
      "step": 34
    },
    {
      "classification_loss": 0.7095,
      "epoch": 0.18543046357615894,
      "grad_norm": 9.728158950805664,
      "learning_rate": 1.2389380530973452e-05,
      "lm_loss": 0.0639,
      "loss": 0.7734,
      "step": 35
    },
    {
      "classification_loss": 0.788,
      "epoch": 0.19072847682119207,
      "grad_norm": 14.292204856872559,
      "learning_rate": 1.2743362831858408e-05,
      "lm_loss": 0.0624,
      "loss": 0.8505,
      "step": 36
    },
    {
      "classification_loss": 0.5976,
      "epoch": 0.19602649006622516,
      "grad_norm": 10.077054023742676,
      "learning_rate": 1.3097345132743363e-05,
      "lm_loss": 0.0666,
      "loss": 0.6642,
      "step": 37
    },
    {
      "classification_loss": 0.8272,
      "epoch": 0.20132450331125828,
      "grad_norm": 10.665119171142578,
      "learning_rate": 1.345132743362832e-05,
      "lm_loss": 0.0827,
      "loss": 0.9098,
      "step": 38
    },
    {
      "classification_loss": 0.8,
      "epoch": 0.20662251655629138,
      "grad_norm": 9.691391944885254,
      "learning_rate": 1.3805309734513275e-05,
      "lm_loss": 0.0902,
      "loss": 0.8902,
      "step": 39
    },
    {
      "classification_loss": 0.6302,
      "epoch": 0.2119205298013245,
      "grad_norm": 8.106285095214844,
      "learning_rate": 1.4159292035398232e-05,
      "lm_loss": 0.0805,
      "loss": 0.7107,
      "step": 40
    },
    {
      "classification_loss": 0.8297,
      "epoch": 0.21721854304635763,
      "grad_norm": 9.286564826965332,
      "learning_rate": 1.4513274336283187e-05,
      "lm_loss": 0.0676,
      "loss": 0.8973,
      "step": 41
    },
    {
      "classification_loss": 0.7143,
      "epoch": 0.22251655629139072,
      "grad_norm": 6.8229289054870605,
      "learning_rate": 1.4867256637168143e-05,
      "lm_loss": 0.0711,
      "loss": 0.7854,
      "step": 42
    },
    {
      "classification_loss": 0.6753,
      "epoch": 0.22781456953642384,
      "grad_norm": 6.802961826324463,
      "learning_rate": 1.5221238938053098e-05,
      "lm_loss": 0.0634,
      "loss": 0.7387,
      "step": 43
    },
    {
      "classification_loss": 0.7426,
      "epoch": 0.23311258278145697,
      "grad_norm": 7.607089996337891,
      "learning_rate": 1.5575221238938054e-05,
      "lm_loss": 0.074,
      "loss": 0.8166,
      "step": 44
    },
    {
      "classification_loss": 0.7181,
      "epoch": 0.23841059602649006,
      "grad_norm": 6.736391067504883,
      "learning_rate": 1.592920353982301e-05,
      "lm_loss": 0.0679,
      "loss": 0.786,
      "step": 45
    },
    {
      "classification_loss": 0.6847,
      "epoch": 0.24370860927152319,
      "grad_norm": 10.298168182373047,
      "learning_rate": 1.628318584070797e-05,
      "lm_loss": 0.0638,
      "loss": 0.7485,
      "step": 46
    },
    {
      "classification_loss": 0.7155,
      "epoch": 0.24900662251655628,
      "grad_norm": 7.87156867980957,
      "learning_rate": 1.663716814159292e-05,
      "lm_loss": 0.0618,
      "loss": 0.7774,
      "step": 47
    },
    {
      "classification_loss": 0.7307,
      "epoch": 0.2543046357615894,
      "grad_norm": 10.80373477935791,
      "learning_rate": 1.6991150442477876e-05,
      "lm_loss": 0.0807,
      "loss": 0.8114,
      "step": 48
    },
    {
      "classification_loss": 0.6405,
      "epoch": 0.2596026490066225,
      "grad_norm": 6.1459221839904785,
      "learning_rate": 1.7345132743362835e-05,
      "lm_loss": 0.073,
      "loss": 0.7135,
      "step": 49
    },
    {
      "classification_loss": 0.6573,
      "epoch": 0.26490066225165565,
      "grad_norm": 7.130291938781738,
      "learning_rate": 1.769911504424779e-05,
      "lm_loss": 0.0713,
      "loss": 0.7286,
      "step": 50
    },
    {
      "epoch": 0.26490066225165565,
      "eval_accuracy": 0.5183098316192627,
      "eval_auroc": 0.6635965704917908,
      "eval_classification_loss": 0.6862303614616394,
      "eval_f1": 0.022857142612338066,
      "eval_lm_loss": 0.06950296461582184,
      "eval_loss": 0.7557332515716553,
      "eval_precision": 1.0,
      "eval_recall": 0.011560693383216858,
      "eval_runtime": 59.1366,
      "eval_samples_per_second": 6.003,
      "eval_steps_per_second": 3.01,
      "step": 50
    },
    {
      "classification_loss": 0.7355,
      "epoch": 0.27019867549668874,
      "grad_norm": 6.673358917236328,
      "learning_rate": 1.8053097345132743e-05,
      "lm_loss": 0.0706,
      "loss": 0.8061,
      "step": 51
    },
    {
      "classification_loss": 0.6574,
      "epoch": 0.27549668874172184,
      "grad_norm": 7.024752140045166,
      "learning_rate": 1.8407079646017702e-05,
      "lm_loss": 0.0738,
      "loss": 0.7311,
      "step": 52
    },
    {
      "classification_loss": 0.7704,
      "epoch": 0.280794701986755,
      "grad_norm": 11.17724895477295,
      "learning_rate": 1.8761061946902657e-05,
      "lm_loss": 0.0793,
      "loss": 0.8496,
      "step": 53
    },
    {
      "classification_loss": 0.7434,
      "epoch": 0.2860927152317881,
      "grad_norm": 6.69635009765625,
      "learning_rate": 1.9115044247787613e-05,
      "lm_loss": 0.0621,
      "loss": 0.8056,
      "step": 54
    },
    {
      "classification_loss": 0.7729,
      "epoch": 0.2913907284768212,
      "grad_norm": 6.636256217956543,
      "learning_rate": 1.946902654867257e-05,
      "lm_loss": 0.0628,
      "loss": 0.8357,
      "step": 55
    },
    {
      "classification_loss": 0.6548,
      "epoch": 0.29668874172185433,
      "grad_norm": 6.512003421783447,
      "learning_rate": 1.9823008849557524e-05,
      "lm_loss": 0.057,
      "loss": 0.7118,
      "step": 56
    },
    {
      "classification_loss": 0.6667,
      "epoch": 0.3019867549668874,
      "grad_norm": 6.662064075469971,
      "learning_rate": 2.017699115044248e-05,
      "lm_loss": 0.0825,
      "loss": 0.7491,
      "step": 57
    },
    {
      "classification_loss": 0.6959,
      "epoch": 0.3072847682119205,
      "grad_norm": 7.244060516357422,
      "learning_rate": 2.0530973451327435e-05,
      "lm_loss": 0.0864,
      "loss": 0.7824,
      "step": 58
    },
    {
      "classification_loss": 0.6951,
      "epoch": 0.3125827814569536,
      "grad_norm": 6.355930805206299,
      "learning_rate": 2.0884955752212394e-05,
      "lm_loss": 0.0663,
      "loss": 0.7614,
      "step": 59
    },
    {
      "classification_loss": 0.7637,
      "epoch": 0.31788079470198677,
      "grad_norm": 7.595324516296387,
      "learning_rate": 2.123893805309735e-05,
      "lm_loss": 0.0722,
      "loss": 0.8359,
      "step": 60
    },
    {
      "classification_loss": 0.6123,
      "epoch": 0.32317880794701986,
      "grad_norm": 5.131961822509766,
      "learning_rate": 2.15929203539823e-05,
      "lm_loss": 0.06,
      "loss": 0.6723,
      "step": 61
    },
    {
      "classification_loss": 0.717,
      "epoch": 0.32847682119205296,
      "grad_norm": 6.761212348937988,
      "learning_rate": 2.1946902654867257e-05,
      "lm_loss": 0.0726,
      "loss": 0.7897,
      "step": 62
    },
    {
      "classification_loss": 0.7763,
      "epoch": 0.3337748344370861,
      "grad_norm": 10.687786102294922,
      "learning_rate": 2.2300884955752213e-05,
      "lm_loss": 0.086,
      "loss": 0.8624,
      "step": 63
    },
    {
      "classification_loss": 0.8011,
      "epoch": 0.3390728476821192,
      "grad_norm": 8.869246482849121,
      "learning_rate": 2.2654867256637168e-05,
      "lm_loss": 0.0716,
      "loss": 0.8726,
      "step": 64
    },
    {
      "classification_loss": 0.7048,
      "epoch": 0.3443708609271523,
      "grad_norm": 9.728758811950684,
      "learning_rate": 2.3008849557522127e-05,
      "lm_loss": 0.0589,
      "loss": 0.7637,
      "step": 65
    },
    {
      "classification_loss": 0.619,
      "epoch": 0.34966887417218545,
      "grad_norm": 5.047206401824951,
      "learning_rate": 2.3362831858407083e-05,
      "lm_loss": 0.0659,
      "loss": 0.6849,
      "step": 66
    },
    {
      "classification_loss": 0.7381,
      "epoch": 0.35496688741721855,
      "grad_norm": 9.180241584777832,
      "learning_rate": 2.3716814159292038e-05,
      "lm_loss": 0.102,
      "loss": 0.84,
      "step": 67
    },
    {
      "classification_loss": 0.6498,
      "epoch": 0.36026490066225164,
      "grad_norm": 5.622615337371826,
      "learning_rate": 2.4070796460176994e-05,
      "lm_loss": 0.0704,
      "loss": 0.7202,
      "step": 68
    },
    {
      "classification_loss": 0.7056,
      "epoch": 0.3655629139072848,
      "grad_norm": 5.238168239593506,
      "learning_rate": 2.4424778761061946e-05,
      "lm_loss": 0.0719,
      "loss": 0.7775,
      "step": 69
    },
    {
      "classification_loss": 0.6488,
      "epoch": 0.3708609271523179,
      "grad_norm": 5.250248908996582,
      "learning_rate": 2.4778761061946905e-05,
      "lm_loss": 0.0607,
      "loss": 0.7095,
      "step": 70
    },
    {
      "classification_loss": 0.7095,
      "epoch": 0.376158940397351,
      "grad_norm": 7.3583292961120605,
      "learning_rate": 2.513274336283186e-05,
      "lm_loss": 0.0709,
      "loss": 0.7805,
      "step": 71
    },
    {
      "classification_loss": 0.6524,
      "epoch": 0.38145695364238413,
      "grad_norm": 6.005558490753174,
      "learning_rate": 2.5486725663716816e-05,
      "lm_loss": 0.0773,
      "loss": 0.7297,
      "step": 72
    },
    {
      "classification_loss": 0.636,
      "epoch": 0.38675496688741723,
      "grad_norm": 5.635666370391846,
      "learning_rate": 2.584070796460177e-05,
      "lm_loss": 0.0576,
      "loss": 0.6936,
      "step": 73
    },
    {
      "classification_loss": 0.66,
      "epoch": 0.3920529801324503,
      "grad_norm": 7.1409711837768555,
      "learning_rate": 2.6194690265486727e-05,
      "lm_loss": 0.0618,
      "loss": 0.7217,
      "step": 74
    },
    {
      "classification_loss": 0.6378,
      "epoch": 0.3973509933774834,
      "grad_norm": 15.095340728759766,
      "learning_rate": 2.6548672566371686e-05,
      "lm_loss": 0.1724,
      "loss": 0.8103,
      "step": 75
    },
    {
      "classification_loss": 0.7661,
      "epoch": 0.40264900662251657,
      "grad_norm": 8.098548889160156,
      "learning_rate": 2.690265486725664e-05,
      "lm_loss": 0.0936,
      "loss": 0.8597,
      "step": 76
    },
    {
      "classification_loss": 0.6327,
      "epoch": 0.40794701986754967,
      "grad_norm": 5.452849388122559,
      "learning_rate": 2.7256637168141594e-05,
      "lm_loss": 0.0596,
      "loss": 0.6923,
      "step": 77
    },
    {
      "classification_loss": 0.6973,
      "epoch": 0.41324503311258276,
      "grad_norm": 8.51659870147705,
      "learning_rate": 2.761061946902655e-05,
      "lm_loss": 0.0748,
      "loss": 0.7721,
      "step": 78
    },
    {
      "classification_loss": 0.741,
      "epoch": 0.4185430463576159,
      "grad_norm": 9.095948219299316,
      "learning_rate": 2.7964601769911505e-05,
      "lm_loss": 0.09,
      "loss": 0.8309,
      "step": 79
    },
    {
      "classification_loss": 0.6963,
      "epoch": 0.423841059602649,
      "grad_norm": 8.077024459838867,
      "learning_rate": 2.8318584070796464e-05,
      "lm_loss": 0.0731,
      "loss": 0.7694,
      "step": 80
    },
    {
      "classification_loss": 0.6749,
      "epoch": 0.4291390728476821,
      "grad_norm": 6.5163984298706055,
      "learning_rate": 2.867256637168142e-05,
      "lm_loss": 0.0623,
      "loss": 0.7372,
      "step": 81
    },
    {
      "classification_loss": 0.7157,
      "epoch": 0.43443708609271525,
      "grad_norm": 10.983697891235352,
      "learning_rate": 2.9026548672566375e-05,
      "lm_loss": 0.0771,
      "loss": 0.7928,
      "step": 82
    },
    {
      "classification_loss": 0.6744,
      "epoch": 0.43973509933774835,
      "grad_norm": 5.42428731918335,
      "learning_rate": 2.938053097345133e-05,
      "lm_loss": 0.0801,
      "loss": 0.7546,
      "step": 83
    },
    {
      "classification_loss": 0.7352,
      "epoch": 0.44503311258278144,
      "grad_norm": 5.746408939361572,
      "learning_rate": 2.9734513274336286e-05,
      "lm_loss": 0.0788,
      "loss": 0.814,
      "step": 84
    },
    {
      "classification_loss": 0.5772,
      "epoch": 0.4503311258278146,
      "grad_norm": 4.765161991119385,
      "learning_rate": 3.0088495575221238e-05,
      "lm_loss": 0.0534,
      "loss": 0.6306,
      "step": 85
    },
    {
      "classification_loss": 0.6584,
      "epoch": 0.4556291390728477,
      "grad_norm": 5.523327827453613,
      "learning_rate": 3.0442477876106197e-05,
      "lm_loss": 0.0696,
      "loss": 0.728,
      "step": 86
    },
    {
      "classification_loss": 0.6058,
      "epoch": 0.4609271523178808,
      "grad_norm": 4.980808258056641,
      "learning_rate": 3.079646017699115e-05,
      "lm_loss": 0.0674,
      "loss": 0.6733,
      "step": 87
    },
    {
      "classification_loss": 0.7551,
      "epoch": 0.46622516556291393,
      "grad_norm": 5.46107816696167,
      "learning_rate": 3.115044247787611e-05,
      "lm_loss": 0.0768,
      "loss": 0.8319,
      "step": 88
    },
    {
      "classification_loss": 0.5283,
      "epoch": 0.47152317880794703,
      "grad_norm": 4.518547534942627,
      "learning_rate": 3.1504424778761064e-05,
      "lm_loss": 0.0596,
      "loss": 0.5879,
      "step": 89
    },
    {
      "classification_loss": 0.72,
      "epoch": 0.4768211920529801,
      "grad_norm": 8.499037742614746,
      "learning_rate": 3.185840707964602e-05,
      "lm_loss": 0.0733,
      "loss": 0.7933,
      "step": 90
    },
    {
      "classification_loss": 0.5866,
      "epoch": 0.4821192052980132,
      "grad_norm": 5.173723220825195,
      "learning_rate": 3.2212389380530975e-05,
      "lm_loss": 0.0611,
      "loss": 0.6478,
      "step": 91
    },
    {
      "classification_loss": 0.5178,
      "epoch": 0.48741721854304637,
      "grad_norm": 4.782351493835449,
      "learning_rate": 3.256637168141594e-05,
      "lm_loss": 0.0543,
      "loss": 0.5721,
      "step": 92
    },
    {
      "classification_loss": 0.5517,
      "epoch": 0.49271523178807947,
      "grad_norm": 6.153548240661621,
      "learning_rate": 3.2920353982300886e-05,
      "lm_loss": 0.0707,
      "loss": 0.6224,
      "step": 93
    },
    {
      "classification_loss": 0.8814,
      "epoch": 0.49801324503311256,
      "grad_norm": 10.506904602050781,
      "learning_rate": 3.327433628318584e-05,
      "lm_loss": 0.0825,
      "loss": 0.9639,
      "step": 94
    },
    {
      "classification_loss": 0.6077,
      "epoch": 0.5033112582781457,
      "grad_norm": 6.529936790466309,
      "learning_rate": 3.36283185840708e-05,
      "lm_loss": 0.0591,
      "loss": 0.6668,
      "step": 95
    },
    {
      "classification_loss": 0.48,
      "epoch": 0.5086092715231788,
      "grad_norm": 7.468279838562012,
      "learning_rate": 3.398230088495575e-05,
      "lm_loss": 0.0491,
      "loss": 0.5291,
      "step": 96
    },
    {
      "classification_loss": 0.5624,
      "epoch": 0.5139072847682119,
      "grad_norm": 7.630971431732178,
      "learning_rate": 3.433628318584071e-05,
      "lm_loss": 0.0549,
      "loss": 0.6173,
      "step": 97
    },
    {
      "classification_loss": 0.4594,
      "epoch": 0.519205298013245,
      "grad_norm": 5.192417621612549,
      "learning_rate": 3.469026548672567e-05,
      "lm_loss": 0.0479,
      "loss": 0.5073,
      "step": 98
    },
    {
      "classification_loss": 0.4582,
      "epoch": 0.5245033112582781,
      "grad_norm": 5.4891743659973145,
      "learning_rate": 3.5044247787610626e-05,
      "lm_loss": 0.0488,
      "loss": 0.507,
      "step": 99
    },
    {
      "classification_loss": 0.8763,
      "epoch": 0.5298013245033113,
      "grad_norm": 13.538164138793945,
      "learning_rate": 3.539823008849558e-05,
      "lm_loss": 0.0962,
      "loss": 0.9725,
      "step": 100
    },
    {
      "epoch": 0.5298013245033113,
      "eval_accuracy": 0.6056337952613831,
      "eval_auroc": 0.7578607201576233,
      "eval_classification_loss": 0.7061096429824829,
      "eval_f1": 0.35779815912246704,
      "eval_lm_loss": 0.0735478326678276,
      "eval_loss": 0.7796574831008911,
      "eval_precision": 0.8666666746139526,
      "eval_recall": 0.22543352842330933,
      "eval_runtime": 59.1917,
      "eval_samples_per_second": 5.997,
      "eval_steps_per_second": 3.007,
      "step": 100
    },
    {
      "classification_loss": 0.7925,
      "epoch": 0.5350993377483444,
      "grad_norm": 10.163474082946777,
      "learning_rate": 3.575221238938053e-05,
      "lm_loss": 0.0672,
      "loss": 0.8597,
      "step": 101
    },
    {
      "classification_loss": 0.6934,
      "epoch": 0.5403973509933775,
      "grad_norm": 9.027915954589844,
      "learning_rate": 3.6106194690265486e-05,
      "lm_loss": 0.0758,
      "loss": 0.7692,
      "step": 102
    },
    {
      "classification_loss": 0.6237,
      "epoch": 0.5456953642384106,
      "grad_norm": 5.972221851348877,
      "learning_rate": 3.646017699115044e-05,
      "lm_loss": 0.0683,
      "loss": 0.6921,
      "step": 103
    },
    {
      "classification_loss": 0.9231,
      "epoch": 0.5509933774834437,
      "grad_norm": 11.96899127960205,
      "learning_rate": 3.6814159292035403e-05,
      "lm_loss": 0.0987,
      "loss": 1.0217,
      "step": 104
    },
    {
      "classification_loss": 0.5976,
      "epoch": 0.5562913907284768,
      "grad_norm": 9.128308296203613,
      "learning_rate": 3.716814159292036e-05,
      "lm_loss": 0.0607,
      "loss": 0.6583,
      "step": 105
    },
    {
      "classification_loss": 0.5443,
      "epoch": 0.56158940397351,
      "grad_norm": 4.956423759460449,
      "learning_rate": 3.7522123893805314e-05,
      "lm_loss": 0.0595,
      "loss": 0.6038,
      "step": 106
    },
    {
      "classification_loss": 0.4808,
      "epoch": 0.5668874172185431,
      "grad_norm": 4.773530006408691,
      "learning_rate": 3.787610619469027e-05,
      "lm_loss": 0.0418,
      "loss": 0.5226,
      "step": 107
    },
    {
      "classification_loss": 0.7282,
      "epoch": 0.5721854304635762,
      "grad_norm": 8.590517044067383,
      "learning_rate": 3.8230088495575226e-05,
      "lm_loss": 0.0632,
      "loss": 0.7914,
      "step": 108
    },
    {
      "classification_loss": 0.5775,
      "epoch": 0.5774834437086093,
      "grad_norm": 9.860918045043945,
      "learning_rate": 3.858407079646018e-05,
      "lm_loss": 0.0405,
      "loss": 0.618,
      "step": 109
    },
    {
      "classification_loss": 0.7147,
      "epoch": 0.5827814569536424,
      "grad_norm": 6.64130163192749,
      "learning_rate": 3.893805309734514e-05,
      "lm_loss": 0.0913,
      "loss": 0.806,
      "step": 110
    },
    {
      "classification_loss": 0.5281,
      "epoch": 0.5880794701986755,
      "grad_norm": 4.763850212097168,
      "learning_rate": 3.929203539823009e-05,
      "lm_loss": 0.0526,
      "loss": 0.5808,
      "step": 111
    },
    {
      "classification_loss": 0.6265,
      "epoch": 0.5933774834437087,
      "grad_norm": 5.150256633758545,
      "learning_rate": 3.964601769911505e-05,
      "lm_loss": 0.0647,
      "loss": 0.6912,
      "step": 112
    },
    {
      "classification_loss": 0.8801,
      "epoch": 0.5986754966887418,
      "grad_norm": 7.241209030151367,
      "learning_rate": 4e-05,
      "lm_loss": 0.1043,
      "loss": 0.9844,
      "step": 113
    },
    {
      "classification_loss": 0.7122,
      "epoch": 0.6039735099337749,
      "grad_norm": 11.587190628051758,
      "learning_rate": 3.999990419960182e-05,
      "lm_loss": 0.0571,
      "loss": 0.7693,
      "step": 114
    },
    {
      "classification_loss": 0.6538,
      "epoch": 0.609271523178808,
      "grad_norm": 6.273010730743408,
      "learning_rate": 3.999961679932504e-05,
      "lm_loss": 0.062,
      "loss": 0.7158,
      "step": 115
    },
    {
      "classification_loss": 0.6977,
      "epoch": 0.614569536423841,
      "grad_norm": 11.858461380004883,
      "learning_rate": 3.999913780192298e-05,
      "lm_loss": 0.0701,
      "loss": 0.7678,
      "step": 116
    },
    {
      "classification_loss": 0.5415,
      "epoch": 0.6198675496688741,
      "grad_norm": 4.724062919616699,
      "learning_rate": 3.999846721198443e-05,
      "lm_loss": 0.0493,
      "loss": 0.5908,
      "step": 117
    },
    {
      "classification_loss": 0.4402,
      "epoch": 0.6251655629139072,
      "grad_norm": 5.963729381561279,
      "learning_rate": 3.9997605035933704e-05,
      "lm_loss": 0.0404,
      "loss": 0.4805,
      "step": 118
    },
    {
      "classification_loss": 0.3184,
      "epoch": 0.6304635761589404,
      "grad_norm": 4.11544942855835,
      "learning_rate": 3.999655128203045e-05,
      "lm_loss": 0.025,
      "loss": 0.3434,
      "step": 119
    },
    {
      "classification_loss": 0.7448,
      "epoch": 0.6357615894039735,
      "grad_norm": 11.775870323181152,
      "learning_rate": 3.999530596036968e-05,
      "lm_loss": 0.0861,
      "loss": 0.8309,
      "step": 120
    },
    {
      "classification_loss": 0.6762,
      "epoch": 0.6410596026490066,
      "grad_norm": 12.822158813476562,
      "learning_rate": 3.9993869082881644e-05,
      "lm_loss": 0.0555,
      "loss": 0.7316,
      "step": 121
    },
    {
      "classification_loss": 0.7036,
      "epoch": 0.6463576158940397,
      "grad_norm": 11.257711410522461,
      "learning_rate": 3.999224066333167e-05,
      "lm_loss": 0.0645,
      "loss": 0.7681,
      "step": 122
    },
    {
      "classification_loss": 0.5842,
      "epoch": 0.6516556291390728,
      "grad_norm": 8.114992141723633,
      "learning_rate": 3.9990420717320074e-05,
      "lm_loss": 0.0666,
      "loss": 0.6508,
      "step": 123
    },
    {
      "classification_loss": 0.3527,
      "epoch": 0.6569536423841059,
      "grad_norm": 8.33047103881836,
      "learning_rate": 3.998840926228203e-05,
      "lm_loss": 0.0366,
      "loss": 0.3893,
      "step": 124
    },
    {
      "classification_loss": 0.5935,
      "epoch": 0.6622516556291391,
      "grad_norm": 7.462849140167236,
      "learning_rate": 3.9986206317487346e-05,
      "lm_loss": 0.065,
      "loss": 0.6585,
      "step": 125
    },
    {
      "classification_loss": 0.3602,
      "epoch": 0.6675496688741722,
      "grad_norm": 7.660101890563965,
      "learning_rate": 3.998381190404032e-05,
      "lm_loss": 0.0423,
      "loss": 0.4025,
      "step": 126
    },
    {
      "classification_loss": 0.8185,
      "epoch": 0.6728476821192053,
      "grad_norm": 10.620638847351074,
      "learning_rate": 3.998122604487954e-05,
      "lm_loss": 0.0641,
      "loss": 0.8826,
      "step": 127
    },
    {
      "classification_loss": 1.1929,
      "epoch": 0.6781456953642384,
      "grad_norm": 23.24151039123535,
      "learning_rate": 3.9978448764777626e-05,
      "lm_loss": 0.0957,
      "loss": 1.2886,
      "step": 128
    },
    {
      "classification_loss": 0.6659,
      "epoch": 0.6834437086092715,
      "grad_norm": 10.195058822631836,
      "learning_rate": 3.997548009034104e-05,
      "lm_loss": 0.0494,
      "loss": 0.7154,
      "step": 129
    },
    {
      "classification_loss": 0.9132,
      "epoch": 0.6887417218543046,
      "grad_norm": 11.151546478271484,
      "learning_rate": 3.997232005000979e-05,
      "lm_loss": 0.0651,
      "loss": 0.9783,
      "step": 130
    },
    {
      "classification_loss": 0.7375,
      "epoch": 0.6940397350993377,
      "grad_norm": 7.32249116897583,
      "learning_rate": 3.996896867405721e-05,
      "lm_loss": 0.0686,
      "loss": 0.8062,
      "step": 131
    },
    {
      "classification_loss": 0.7029,
      "epoch": 0.6993377483443709,
      "grad_norm": 7.1324663162231445,
      "learning_rate": 3.9965425994589594e-05,
      "lm_loss": 0.0695,
      "loss": 0.7724,
      "step": 132
    },
    {
      "classification_loss": 0.6562,
      "epoch": 0.704635761589404,
      "grad_norm": 5.124715805053711,
      "learning_rate": 3.9961692045545956e-05,
      "lm_loss": 0.08,
      "loss": 0.7362,
      "step": 133
    },
    {
      "classification_loss": 0.5983,
      "epoch": 0.7099337748344371,
      "grad_norm": 4.121169567108154,
      "learning_rate": 3.995776686269769e-05,
      "lm_loss": 0.1008,
      "loss": 0.6991,
      "step": 134
    },
    {
      "classification_loss": 0.5676,
      "epoch": 0.7152317880794702,
      "grad_norm": 3.843527317047119,
      "learning_rate": 3.995365048364819e-05,
      "lm_loss": 0.0836,
      "loss": 0.6512,
      "step": 135
    },
    {
      "classification_loss": 0.6631,
      "epoch": 0.7205298013245033,
      "grad_norm": 5.859713554382324,
      "learning_rate": 3.994934294783254e-05,
      "lm_loss": 0.0691,
      "loss": 0.7322,
      "step": 136
    },
    {
      "classification_loss": 0.6996,
      "epoch": 0.7258278145695364,
      "grad_norm": 7.880208969116211,
      "learning_rate": 3.99448442965171e-05,
      "lm_loss": 0.058,
      "loss": 0.7576,
      "step": 137
    },
    {
      "classification_loss": 0.6612,
      "epoch": 0.7311258278145696,
      "grad_norm": 6.112351894378662,
      "learning_rate": 3.994015457279912e-05,
      "lm_loss": 0.0671,
      "loss": 0.7283,
      "step": 138
    },
    {
      "classification_loss": 0.5654,
      "epoch": 0.7364238410596027,
      "grad_norm": 5.5251641273498535,
      "learning_rate": 3.993527382160636e-05,
      "lm_loss": 0.0511,
      "loss": 0.6165,
      "step": 139
    },
    {
      "classification_loss": 0.5681,
      "epoch": 0.7417218543046358,
      "grad_norm": 4.911803245544434,
      "learning_rate": 3.99302020896966e-05,
      "lm_loss": 0.0624,
      "loss": 0.6305,
      "step": 140
    },
    {
      "classification_loss": 0.4893,
      "epoch": 0.7470198675496689,
      "grad_norm": 6.811244487762451,
      "learning_rate": 3.9924939425657225e-05,
      "lm_loss": 0.0344,
      "loss": 0.5236,
      "step": 141
    },
    {
      "classification_loss": 0.5964,
      "epoch": 0.752317880794702,
      "grad_norm": 5.395686149597168,
      "learning_rate": 3.991948587990479e-05,
      "lm_loss": 0.0605,
      "loss": 0.6569,
      "step": 142
    },
    {
      "classification_loss": 0.4877,
      "epoch": 0.7576158940397351,
      "grad_norm": 4.27349328994751,
      "learning_rate": 3.991384150468445e-05,
      "lm_loss": 0.0372,
      "loss": 0.5249,
      "step": 143
    },
    {
      "classification_loss": 0.6091,
      "epoch": 0.7629139072847683,
      "grad_norm": 6.668065071105957,
      "learning_rate": 3.9908006354069565e-05,
      "lm_loss": 0.0688,
      "loss": 0.6779,
      "step": 144
    },
    {
      "classification_loss": 0.8226,
      "epoch": 0.7682119205298014,
      "grad_norm": 10.073720932006836,
      "learning_rate": 3.990198048396111e-05,
      "lm_loss": 0.0945,
      "loss": 0.9171,
      "step": 145
    },
    {
      "classification_loss": 0.5477,
      "epoch": 0.7735099337748345,
      "grad_norm": 6.441249847412109,
      "learning_rate": 3.989576395208715e-05,
      "lm_loss": 0.0637,
      "loss": 0.6114,
      "step": 146
    },
    {
      "classification_loss": 0.5658,
      "epoch": 0.7788079470198676,
      "grad_norm": 9.964797973632812,
      "learning_rate": 3.988935681800231e-05,
      "lm_loss": 0.062,
      "loss": 0.6277,
      "step": 147
    },
    {
      "classification_loss": 0.4358,
      "epoch": 0.7841059602649006,
      "grad_norm": 6.116605758666992,
      "learning_rate": 3.9882759143087194e-05,
      "lm_loss": 0.0439,
      "loss": 0.4797,
      "step": 148
    },
    {
      "classification_loss": 0.8347,
      "epoch": 0.7894039735099337,
      "grad_norm": 9.535455703735352,
      "learning_rate": 3.987597099054779e-05,
      "lm_loss": 0.1175,
      "loss": 0.9522,
      "step": 149
    },
    {
      "classification_loss": 0.5125,
      "epoch": 0.7947019867549668,
      "grad_norm": 8.548845291137695,
      "learning_rate": 3.9868992425414876e-05,
      "lm_loss": 0.0458,
      "loss": 0.5584,
      "step": 150
    },
    {
      "epoch": 0.7947019867549668,
      "eval_accuracy": 0.7154929637908936,
      "eval_auroc": 0.8118211030960083,
      "eval_classification_loss": 0.530834436416626,
      "eval_f1": 0.6833855509757996,
      "eval_lm_loss": 0.060030627995729446,
      "eval_loss": 0.5908650755882263,
      "eval_precision": 0.7465753555297852,
      "eval_recall": 0.6300578117370605,
      "eval_runtime": 63.9615,
      "eval_samples_per_second": 5.55,
      "eval_steps_per_second": 2.783,
      "step": 150
    },
    {
      "classification_loss": 0.4766,
      "epoch": 0.8,
      "grad_norm": 9.953009605407715,
      "learning_rate": 3.986182351454337e-05,
      "lm_loss": 0.0471,
      "loss": 0.5237,
      "step": 151
    },
    {
      "classification_loss": 0.5022,
      "epoch": 0.8052980132450331,
      "grad_norm": 7.446337699890137,
      "learning_rate": 3.9854464326611735e-05,
      "lm_loss": 0.0386,
      "loss": 0.5408,
      "step": 152
    },
    {
      "classification_loss": 0.6631,
      "epoch": 0.8105960264900662,
      "grad_norm": 16.3365535736084,
      "learning_rate": 3.9846914932121274e-05,
      "lm_loss": 0.0717,
      "loss": 0.7348,
      "step": 153
    },
    {
      "classification_loss": 0.7707,
      "epoch": 0.8158940397350993,
      "grad_norm": 19.081138610839844,
      "learning_rate": 3.983917540339549e-05,
      "lm_loss": 0.087,
      "loss": 0.8578,
      "step": 154
    },
    {
      "classification_loss": 0.5546,
      "epoch": 0.8211920529801324,
      "grad_norm": 7.790853500366211,
      "learning_rate": 3.983124581457939e-05,
      "lm_loss": 0.0475,
      "loss": 0.6021,
      "step": 155
    },
    {
      "classification_loss": 0.5297,
      "epoch": 0.8264900662251655,
      "grad_norm": 10.940109252929688,
      "learning_rate": 3.982312624163874e-05,
      "lm_loss": 0.06,
      "loss": 0.5897,
      "step": 156
    },
    {
      "classification_loss": 0.5669,
      "epoch": 0.8317880794701987,
      "grad_norm": 6.8043131828308105,
      "learning_rate": 3.981481676235937e-05,
      "lm_loss": 0.0556,
      "loss": 0.6225,
      "step": 157
    },
    {
      "classification_loss": 0.5921,
      "epoch": 0.8370860927152318,
      "grad_norm": 9.720118522644043,
      "learning_rate": 3.980631745634641e-05,
      "lm_loss": 0.0628,
      "loss": 0.6549,
      "step": 158
    },
    {
      "classification_loss": 0.4736,
      "epoch": 0.8423841059602649,
      "grad_norm": 5.871112823486328,
      "learning_rate": 3.979762840502358e-05,
      "lm_loss": 0.0413,
      "loss": 0.5149,
      "step": 159
    },
    {
      "classification_loss": 0.6658,
      "epoch": 0.847682119205298,
      "grad_norm": 7.333876609802246,
      "learning_rate": 3.978874969163231e-05,
      "lm_loss": 0.0731,
      "loss": 0.7389,
      "step": 160
    },
    {
      "classification_loss": 0.5645,
      "epoch": 0.8529801324503311,
      "grad_norm": 9.882183074951172,
      "learning_rate": 3.9779681401231046e-05,
      "lm_loss": 0.0506,
      "loss": 0.615,
      "step": 161
    },
    {
      "classification_loss": 0.505,
      "epoch": 0.8582781456953642,
      "grad_norm": 8.793054580688477,
      "learning_rate": 3.977042362069437e-05,
      "lm_loss": 0.0492,
      "loss": 0.5543,
      "step": 162
    },
    {
      "classification_loss": 0.5705,
      "epoch": 0.8635761589403973,
      "grad_norm": 7.339094161987305,
      "learning_rate": 3.976097643871217e-05,
      "lm_loss": 0.0655,
      "loss": 0.6361,
      "step": 163
    },
    {
      "classification_loss": 0.4228,
      "epoch": 0.8688741721854305,
      "grad_norm": 5.383102893829346,
      "learning_rate": 3.975133994578885e-05,
      "lm_loss": 0.0518,
      "loss": 0.4746,
      "step": 164
    },
    {
      "classification_loss": 0.4319,
      "epoch": 0.8741721854304636,
      "grad_norm": 5.786334991455078,
      "learning_rate": 3.9741514234242385e-05,
      "lm_loss": 0.0472,
      "loss": 0.4792,
      "step": 165
    },
    {
      "classification_loss": 0.6828,
      "epoch": 0.8794701986754967,
      "grad_norm": 10.55814266204834,
      "learning_rate": 3.973149939820348e-05,
      "lm_loss": 0.0645,
      "loss": 0.7472,
      "step": 166
    },
    {
      "classification_loss": 0.2304,
      "epoch": 0.8847682119205298,
      "grad_norm": 4.401240348815918,
      "learning_rate": 3.972129553361465e-05,
      "lm_loss": 0.0231,
      "loss": 0.2535,
      "step": 167
    },
    {
      "classification_loss": 0.9051,
      "epoch": 0.8900662251655629,
      "grad_norm": 13.669878959655762,
      "learning_rate": 3.9710902738229354e-05,
      "lm_loss": 0.0733,
      "loss": 0.9784,
      "step": 168
    },
    {
      "classification_loss": 0.6184,
      "epoch": 0.895364238410596,
      "grad_norm": 8.987906455993652,
      "learning_rate": 3.970032111161097e-05,
      "lm_loss": 0.0511,
      "loss": 0.6695,
      "step": 169
    },
    {
      "classification_loss": 0.3852,
      "epoch": 0.9006622516556292,
      "grad_norm": 9.206149101257324,
      "learning_rate": 3.96895507551319e-05,
      "lm_loss": 0.0291,
      "loss": 0.4143,
      "step": 170
    },
    {
      "classification_loss": 0.4101,
      "epoch": 0.9059602649006623,
      "grad_norm": 6.79190731048584,
      "learning_rate": 3.96785917719726e-05,
      "lm_loss": 0.0461,
      "loss": 0.4563,
      "step": 171
    },
    {
      "classification_loss": 0.4847,
      "epoch": 0.9112582781456954,
      "grad_norm": 10.307087898254395,
      "learning_rate": 3.966744426712055e-05,
      "lm_loss": 0.0536,
      "loss": 0.5383,
      "step": 172
    },
    {
      "classification_loss": 0.4254,
      "epoch": 0.9165562913907285,
      "grad_norm": 5.463620662689209,
      "learning_rate": 3.965610834736929e-05,
      "lm_loss": 0.0413,
      "loss": 0.4668,
      "step": 173
    },
    {
      "classification_loss": 0.4826,
      "epoch": 0.9218543046357616,
      "grad_norm": 7.979359149932861,
      "learning_rate": 3.96445841213174e-05,
      "lm_loss": 0.0555,
      "loss": 0.5381,
      "step": 174
    },
    {
      "classification_loss": 0.7103,
      "epoch": 0.9271523178807947,
      "grad_norm": 10.336874961853027,
      "learning_rate": 3.963287169936741e-05,
      "lm_loss": 0.0726,
      "loss": 0.7829,
      "step": 175
    },
    {
      "classification_loss": 0.2074,
      "epoch": 0.9324503311258279,
      "grad_norm": 4.121266841888428,
      "learning_rate": 3.962097119372478e-05,
      "lm_loss": 0.0206,
      "loss": 0.228,
      "step": 176
    },
    {
      "classification_loss": 0.6891,
      "epoch": 0.937748344370861,
      "grad_norm": 10.487539291381836,
      "learning_rate": 3.960888271839686e-05,
      "lm_loss": 0.0734,
      "loss": 0.7624,
      "step": 177
    },
    {
      "classification_loss": 0.2912,
      "epoch": 0.9430463576158941,
      "grad_norm": 8.726000785827637,
      "learning_rate": 3.959660638919169e-05,
      "lm_loss": 0.0321,
      "loss": 0.3234,
      "step": 178
    },
    {
      "classification_loss": 0.5447,
      "epoch": 0.9483443708609272,
      "grad_norm": 9.660502433776855,
      "learning_rate": 3.958414232371702e-05,
      "lm_loss": 0.0586,
      "loss": 0.6033,
      "step": 179
    },
    {
      "classification_loss": 0.2764,
      "epoch": 0.9536423841059603,
      "grad_norm": 7.710514545440674,
      "learning_rate": 3.957149064137908e-05,
      "lm_loss": 0.0213,
      "loss": 0.2976,
      "step": 180
    },
    {
      "classification_loss": 0.4994,
      "epoch": 0.9589403973509933,
      "grad_norm": 9.137280464172363,
      "learning_rate": 3.955865146338148e-05,
      "lm_loss": 0.0505,
      "loss": 0.5499,
      "step": 181
    },
    {
      "classification_loss": 0.2702,
      "epoch": 0.9642384105960264,
      "grad_norm": 4.201053142547607,
      "learning_rate": 3.954562491272408e-05,
      "lm_loss": 0.0274,
      "loss": 0.2976,
      "step": 182
    },
    {
      "classification_loss": 0.6516,
      "epoch": 0.9695364238410596,
      "grad_norm": 9.431659698486328,
      "learning_rate": 3.953241111420174e-05,
      "lm_loss": 0.066,
      "loss": 0.7176,
      "step": 183
    },
    {
      "classification_loss": 0.3133,
      "epoch": 0.9748344370860927,
      "grad_norm": 7.765235900878906,
      "learning_rate": 3.9519010194403175e-05,
      "lm_loss": 0.0346,
      "loss": 0.348,
      "step": 184
    },
    {
      "classification_loss": 0.2974,
      "epoch": 0.9801324503311258,
      "grad_norm": 5.19456148147583,
      "learning_rate": 3.950542228170974e-05,
      "lm_loss": 0.0358,
      "loss": 0.3332,
      "step": 185
    },
    {
      "classification_loss": 0.7412,
      "epoch": 0.9854304635761589,
      "grad_norm": 8.688366889953613,
      "learning_rate": 3.9491647506294166e-05,
      "lm_loss": 0.0839,
      "loss": 0.8252,
      "step": 186
    },
    {
      "classification_loss": 0.33,
      "epoch": 0.990728476821192,
      "grad_norm": 5.653778076171875,
      "learning_rate": 3.9477686000119355e-05,
      "lm_loss": 0.037,
      "loss": 0.367,
      "step": 187
    },
    {
      "classification_loss": 0.2157,
      "epoch": 0.9960264900662251,
      "grad_norm": 4.750575542449951,
      "learning_rate": 3.94635378969371e-05,
      "lm_loss": 0.0227,
      "loss": 0.2383,
      "step": 188
    },
    {
      "classification_loss": 0.5733,
      "epoch": 1.0013245033112583,
      "grad_norm": 7.4247846603393555,
      "learning_rate": 3.9449203332286787e-05,
      "lm_loss": 0.05,
      "loss": 0.6233,
      "step": 189
    },
    {
      "classification_loss": 0.3076,
      "epoch": 1.0066225165562914,
      "grad_norm": 3.8869001865386963,
      "learning_rate": 3.943468244349411e-05,
      "lm_loss": 0.0404,
      "loss": 0.3479,
      "step": 190
    },
    {
      "classification_loss": 0.4149,
      "epoch": 1.0119205298013245,
      "grad_norm": 5.648099422454834,
      "learning_rate": 3.941997536966977e-05,
      "lm_loss": 0.0395,
      "loss": 0.4544,
      "step": 191
    },
    {
      "classification_loss": 0.1995,
      "epoch": 1.0172185430463576,
      "grad_norm": 4.823829174041748,
      "learning_rate": 3.940508225170813e-05,
      "lm_loss": 0.02,
      "loss": 0.2195,
      "step": 192
    },
    {
      "classification_loss": 0.3768,
      "epoch": 1.0225165562913907,
      "grad_norm": 6.221853256225586,
      "learning_rate": 3.939000323228583e-05,
      "lm_loss": 0.0351,
      "loss": 0.4119,
      "step": 193
    },
    {
      "classification_loss": 0.2282,
      "epoch": 1.0278145695364238,
      "grad_norm": 3.405147075653076,
      "learning_rate": 3.937473845586049e-05,
      "lm_loss": 0.0221,
      "loss": 0.2504,
      "step": 194
    },
    {
      "classification_loss": 0.2622,
      "epoch": 1.033112582781457,
      "grad_norm": 3.066460609436035,
      "learning_rate": 3.935928806866928e-05,
      "lm_loss": 0.0298,
      "loss": 0.292,
      "step": 195
    },
    {
      "classification_loss": 0.131,
      "epoch": 1.03841059602649,
      "grad_norm": 3.0996804237365723,
      "learning_rate": 3.934365221872751e-05,
      "lm_loss": 0.0139,
      "loss": 0.1449,
      "step": 196
    },
    {
      "classification_loss": 0.6482,
      "epoch": 1.043708609271523,
      "grad_norm": 12.585487365722656,
      "learning_rate": 3.9327831055827255e-05,
      "lm_loss": 0.0704,
      "loss": 0.7185,
      "step": 197
    },
    {
      "classification_loss": 0.7351,
      "epoch": 1.0490066225165562,
      "grad_norm": 12.006387710571289,
      "learning_rate": 3.931182473153588e-05,
      "lm_loss": 0.0657,
      "loss": 0.8008,
      "step": 198
    },
    {
      "classification_loss": 0.179,
      "epoch": 1.0543046357615895,
      "grad_norm": 6.926703929901123,
      "learning_rate": 3.9295633399194614e-05,
      "lm_loss": 0.0142,
      "loss": 0.1932,
      "step": 199
    },
    {
      "classification_loss": 0.5921,
      "epoch": 1.0596026490066226,
      "grad_norm": 13.085765838623047,
      "learning_rate": 3.927925721391707e-05,
      "lm_loss": 0.0483,
      "loss": 0.6404,
      "step": 200
    },
    {
      "epoch": 1.0596026490066226,
      "eval_accuracy": 0.800000011920929,
      "eval_auroc": 0.8991297483444214,
      "eval_classification_loss": 0.5863198041915894,
      "eval_f1": 0.817480742931366,
      "eval_lm_loss": 0.050506215542554855,
      "eval_loss": 0.6368260383605957,
      "eval_precision": 0.7361111044883728,
      "eval_recall": 0.9190751314163208,
      "eval_runtime": 61.0811,
      "eval_samples_per_second": 5.812,
      "eval_steps_per_second": 2.914,
      "step": 200
    },
    {
      "classification_loss": 0.3162,
      "epoch": 1.0649006622516557,
      "grad_norm": 6.91673469543457,
      "learning_rate": 3.926269633258774e-05,
      "lm_loss": 0.022,
      "loss": 0.3381,
      "step": 201
    },
    {
      "classification_loss": 0.4733,
      "epoch": 1.0701986754966888,
      "grad_norm": 8.439972877502441,
      "learning_rate": 3.924595091386055e-05,
      "lm_loss": 0.0414,
      "loss": 0.5147,
      "step": 202
    },
    {
      "classification_loss": 0.1905,
      "epoch": 1.0754966887417219,
      "grad_norm": 6.084334850311279,
      "learning_rate": 3.922902111815725e-05,
      "lm_loss": 0.019,
      "loss": 0.2095,
      "step": 203
    },
    {
      "classification_loss": 0.1698,
      "epoch": 1.080794701986755,
      "grad_norm": 12.098930358886719,
      "learning_rate": 3.9211907107665974e-05,
      "lm_loss": 0.0201,
      "loss": 0.1899,
      "step": 204
    },
    {
      "classification_loss": 0.1284,
      "epoch": 1.086092715231788,
      "grad_norm": 3.2182488441467285,
      "learning_rate": 3.9194609046339625e-05,
      "lm_loss": 0.0124,
      "loss": 0.1409,
      "step": 205
    },
    {
      "classification_loss": 0.1812,
      "epoch": 1.0913907284768212,
      "grad_norm": 8.505109786987305,
      "learning_rate": 3.917712709989432e-05,
      "lm_loss": 0.0227,
      "loss": 0.2039,
      "step": 206
    },
    {
      "classification_loss": 0.3735,
      "epoch": 1.0966887417218543,
      "grad_norm": 7.735908031463623,
      "learning_rate": 3.9159461435807786e-05,
      "lm_loss": 0.0452,
      "loss": 0.4187,
      "step": 207
    },
    {
      "classification_loss": 0.1704,
      "epoch": 1.1019867549668874,
      "grad_norm": 8.231406211853027,
      "learning_rate": 3.91416122233178e-05,
      "lm_loss": 0.0187,
      "loss": 0.1892,
      "step": 208
    },
    {
      "classification_loss": 0.4376,
      "epoch": 1.1072847682119205,
      "grad_norm": 5.866336822509766,
      "learning_rate": 3.912357963342054e-05,
      "lm_loss": 0.0475,
      "loss": 0.4851,
      "step": 209
    },
    {
      "classification_loss": 0.1993,
      "epoch": 1.1125827814569536,
      "grad_norm": 5.63424015045166,
      "learning_rate": 3.910536383886891e-05,
      "lm_loss": 0.0249,
      "loss": 0.2242,
      "step": 210
    },
    {
      "classification_loss": 0.2694,
      "epoch": 1.1178807947019869,
      "grad_norm": 7.457085132598877,
      "learning_rate": 3.908696501417097e-05,
      "lm_loss": 0.0294,
      "loss": 0.2988,
      "step": 211
    },
    {
      "classification_loss": 0.276,
      "epoch": 1.12317880794702,
      "grad_norm": 6.52131462097168,
      "learning_rate": 3.906838333558817e-05,
      "lm_loss": 0.024,
      "loss": 0.2999,
      "step": 212
    },
    {
      "classification_loss": 0.4529,
      "epoch": 1.128476821192053,
      "grad_norm": 9.474043846130371,
      "learning_rate": 3.9049618981133756e-05,
      "lm_loss": 0.0447,
      "loss": 0.4977,
      "step": 213
    },
    {
      "classification_loss": 0.1275,
      "epoch": 1.1337748344370862,
      "grad_norm": 5.4531121253967285,
      "learning_rate": 3.903067213057098e-05,
      "lm_loss": 0.0135,
      "loss": 0.141,
      "step": 214
    },
    {
      "classification_loss": 0.0974,
      "epoch": 1.1390728476821192,
      "grad_norm": 3.256362199783325,
      "learning_rate": 3.901154296541142e-05,
      "lm_loss": 0.0124,
      "loss": 0.1098,
      "step": 215
    },
    {
      "classification_loss": 0.7976,
      "epoch": 1.1443708609271523,
      "grad_norm": 13.55454158782959,
      "learning_rate": 3.899223166891324e-05,
      "lm_loss": 0.0805,
      "loss": 0.8781,
      "step": 216
    },
    {
      "classification_loss": 0.3109,
      "epoch": 1.1496688741721854,
      "grad_norm": 10.241846084594727,
      "learning_rate": 3.897273842607944e-05,
      "lm_loss": 0.0305,
      "loss": 0.3414,
      "step": 217
    },
    {
      "classification_loss": 0.096,
      "epoch": 1.1549668874172185,
      "grad_norm": 4.802555561065674,
      "learning_rate": 3.8953063423656055e-05,
      "lm_loss": 0.0105,
      "loss": 0.1065,
      "step": 218
    },
    {
      "classification_loss": 0.1617,
      "epoch": 1.1602649006622516,
      "grad_norm": 9.075227737426758,
      "learning_rate": 3.893320685013039e-05,
      "lm_loss": 0.016,
      "loss": 0.1777,
      "step": 219
    },
    {
      "classification_loss": 0.2021,
      "epoch": 1.1655629139072847,
      "grad_norm": 14.641780853271484,
      "learning_rate": 3.8913168895729215e-05,
      "lm_loss": 0.0203,
      "loss": 0.2225,
      "step": 220
    },
    {
      "classification_loss": 0.094,
      "epoch": 1.1708609271523178,
      "grad_norm": 6.224979877471924,
      "learning_rate": 3.889294975241692e-05,
      "lm_loss": 0.0087,
      "loss": 0.1027,
      "step": 221
    },
    {
      "classification_loss": 0.2566,
      "epoch": 1.176158940397351,
      "grad_norm": 4.113605976104736,
      "learning_rate": 3.887254961389372e-05,
      "lm_loss": 0.0259,
      "loss": 0.2826,
      "step": 222
    },
    {
      "classification_loss": 0.3488,
      "epoch": 1.181456953642384,
      "grad_norm": 30.50343894958496,
      "learning_rate": 3.885196867559374e-05,
      "lm_loss": 0.0411,
      "loss": 0.3899,
      "step": 223
    },
    {
      "classification_loss": 0.1624,
      "epoch": 1.1867549668874173,
      "grad_norm": 8.688823699951172,
      "learning_rate": 3.88312071346832e-05,
      "lm_loss": 0.0187,
      "loss": 0.1811,
      "step": 224
    },
    {
      "classification_loss": 0.3277,
      "epoch": 1.1920529801324504,
      "grad_norm": 11.881219863891602,
      "learning_rate": 3.881026519005848e-05,
      "lm_loss": 0.043,
      "loss": 0.3707,
      "step": 225
    },
    {
      "classification_loss": 0.1595,
      "epoch": 1.1973509933774835,
      "grad_norm": 77.27059173583984,
      "learning_rate": 3.878914304234425e-05,
      "lm_loss": 0.0165,
      "loss": 0.1759,
      "step": 226
    },
    {
      "classification_loss": 0.1257,
      "epoch": 1.2026490066225166,
      "grad_norm": 6.8302531242370605,
      "learning_rate": 3.876784089389152e-05,
      "lm_loss": 0.0165,
      "loss": 0.1421,
      "step": 227
    },
    {
      "classification_loss": 0.2511,
      "epoch": 1.2079470198675497,
      "grad_norm": 6.045069694519043,
      "learning_rate": 3.8746358948775716e-05,
      "lm_loss": 0.0275,
      "loss": 0.2787,
      "step": 228
    },
    {
      "classification_loss": 0.2979,
      "epoch": 1.2132450331125828,
      "grad_norm": 5.721494197845459,
      "learning_rate": 3.872469741279475e-05,
      "lm_loss": 0.0198,
      "loss": 0.3177,
      "step": 229
    },
    {
      "classification_loss": 0.1741,
      "epoch": 1.218543046357616,
      "grad_norm": 14.820671081542969,
      "learning_rate": 3.870285649346697e-05,
      "lm_loss": 0.0195,
      "loss": 0.1936,
      "step": 230
    },
    {
      "classification_loss": 0.265,
      "epoch": 1.223841059602649,
      "grad_norm": 10.438268661499023,
      "learning_rate": 3.8680836400029276e-05,
      "lm_loss": 0.023,
      "loss": 0.288,
      "step": 231
    },
    {
      "classification_loss": 0.5519,
      "epoch": 1.229139072847682,
      "grad_norm": 9.363056182861328,
      "learning_rate": 3.865863734343502e-05,
      "lm_loss": 0.0606,
      "loss": 0.6125,
      "step": 232
    },
    {
      "classification_loss": 0.256,
      "epoch": 1.2344370860927152,
      "grad_norm": 9.694531440734863,
      "learning_rate": 3.8636259536352066e-05,
      "lm_loss": 0.0222,
      "loss": 0.2782,
      "step": 233
    },
    {
      "classification_loss": 0.2424,
      "epoch": 1.2397350993377483,
      "grad_norm": 4.9929399490356445,
      "learning_rate": 3.861370319316069e-05,
      "lm_loss": 0.0262,
      "loss": 0.2686,
      "step": 234
    },
    {
      "classification_loss": 0.1779,
      "epoch": 1.2450331125827814,
      "grad_norm": 8.662088394165039,
      "learning_rate": 3.859096852995155e-05,
      "lm_loss": 0.0156,
      "loss": 0.1936,
      "step": 235
    },
    {
      "classification_loss": 0.3807,
      "epoch": 1.2503311258278145,
      "grad_norm": 7.348361968994141,
      "learning_rate": 3.856805576452364e-05,
      "lm_loss": 0.031,
      "loss": 0.4117,
      "step": 236
    },
    {
      "classification_loss": 0.4118,
      "epoch": 1.2556291390728478,
      "grad_norm": 12.156649589538574,
      "learning_rate": 3.8544965116382146e-05,
      "lm_loss": 0.0474,
      "loss": 0.4593,
      "step": 237
    },
    {
      "classification_loss": 0.1247,
      "epoch": 1.2609271523178807,
      "grad_norm": 7.528282165527344,
      "learning_rate": 3.852169680673642e-05,
      "lm_loss": 0.0141,
      "loss": 0.1387,
      "step": 238
    },
    {
      "classification_loss": 0.1317,
      "epoch": 1.266225165562914,
      "grad_norm": 8.301989555358887,
      "learning_rate": 3.849825105849778e-05,
      "lm_loss": 0.0107,
      "loss": 0.1424,
      "step": 239
    },
    {
      "classification_loss": 0.1389,
      "epoch": 1.271523178807947,
      "grad_norm": 6.220516204833984,
      "learning_rate": 3.847462809627743e-05,
      "lm_loss": 0.0115,
      "loss": 0.1504,
      "step": 240
    },
    {
      "classification_loss": 0.2853,
      "epoch": 1.2768211920529802,
      "grad_norm": 5.342477798461914,
      "learning_rate": 3.8450828146384286e-05,
      "lm_loss": 0.0337,
      "loss": 0.319,
      "step": 241
    },
    {
      "classification_loss": 0.1328,
      "epoch": 1.2821192052980133,
      "grad_norm": 3.2354812622070312,
      "learning_rate": 3.842685143682282e-05,
      "lm_loss": 0.0123,
      "loss": 0.1451,
      "step": 242
    },
    {
      "classification_loss": 0.2108,
      "epoch": 1.2874172185430464,
      "grad_norm": 5.4710540771484375,
      "learning_rate": 3.8402698197290865e-05,
      "lm_loss": 0.0244,
      "loss": 0.2352,
      "step": 243
    },
    {
      "classification_loss": 0.2048,
      "epoch": 1.2927152317880795,
      "grad_norm": 8.4066743850708,
      "learning_rate": 3.8378368659177415e-05,
      "lm_loss": 0.0206,
      "loss": 0.2254,
      "step": 244
    },
    {
      "classification_loss": 0.3189,
      "epoch": 1.2980132450331126,
      "grad_norm": 7.102217197418213,
      "learning_rate": 3.835386305556042e-05,
      "lm_loss": 0.0323,
      "loss": 0.3512,
      "step": 245
    },
    {
      "classification_loss": 0.3271,
      "epoch": 1.3033112582781456,
      "grad_norm": 8.524945259094238,
      "learning_rate": 3.8329181621204524e-05,
      "lm_loss": 0.0251,
      "loss": 0.3522,
      "step": 246
    },
    {
      "classification_loss": 0.2591,
      "epoch": 1.3086092715231787,
      "grad_norm": 11.300804138183594,
      "learning_rate": 3.8304324592558863e-05,
      "lm_loss": 0.024,
      "loss": 0.2831,
      "step": 247
    },
    {
      "classification_loss": 0.2957,
      "epoch": 1.313907284768212,
      "grad_norm": 8.190695762634277,
      "learning_rate": 3.8279292207754755e-05,
      "lm_loss": 0.0298,
      "loss": 0.3256,
      "step": 248
    },
    {
      "classification_loss": 0.0523,
      "epoch": 1.319205298013245,
      "grad_norm": 1.8362592458724976,
      "learning_rate": 3.825408470660345e-05,
      "lm_loss": 0.0054,
      "loss": 0.0577,
      "step": 249
    },
    {
      "classification_loss": 0.5572,
      "epoch": 1.3245033112582782,
      "grad_norm": 18.345380783081055,
      "learning_rate": 3.822870233059381e-05,
      "lm_loss": 0.0573,
      "loss": 0.6145,
      "step": 250
    },
    {
      "epoch": 1.3245033112582782,
      "eval_accuracy": 0.8197183012962341,
      "eval_auroc": 0.9123419523239136,
      "eval_classification_loss": 0.5566179156303406,
      "eval_f1": 0.8341968655586243,
      "eval_lm_loss": 0.05854928493499756,
      "eval_loss": 0.6151672601699829,
      "eval_precision": 0.7558685541152954,
      "eval_recall": 0.9306358098983765,
      "eval_runtime": 60.2799,
      "eval_samples_per_second": 5.889,
      "eval_steps_per_second": 2.953,
      "step": 250
    },
    {
      "classification_loss": 0.0955,
      "epoch": 1.3298013245033111,
      "grad_norm": 9.865626335144043,
      "learning_rate": 3.8203145322890006e-05,
      "lm_loss": 0.0128,
      "loss": 0.1083,
      "step": 251
    },
    {
      "classification_loss": 0.0311,
      "epoch": 1.3350993377483444,
      "grad_norm": 1.8209079504013062,
      "learning_rate": 3.817741392832919e-05,
      "lm_loss": 0.0034,
      "loss": 0.0344,
      "step": 252
    },
    {
      "classification_loss": 0.4761,
      "epoch": 1.3403973509933775,
      "grad_norm": 8.50853443145752,
      "learning_rate": 3.815150839341915e-05,
      "lm_loss": 0.0606,
      "loss": 0.5367,
      "step": 253
    },
    {
      "classification_loss": 0.5001,
      "epoch": 1.3456953642384106,
      "grad_norm": 11.227840423583984,
      "learning_rate": 3.8125428966335934e-05,
      "lm_loss": 0.0546,
      "loss": 0.5547,
      "step": 254
    },
    {
      "classification_loss": 0.2367,
      "epoch": 1.3509933774834437,
      "grad_norm": 12.700416564941406,
      "learning_rate": 3.8099175896921495e-05,
      "lm_loss": 0.0269,
      "loss": 0.2636,
      "step": 255
    },
    {
      "classification_loss": 0.038,
      "epoch": 1.3562913907284768,
      "grad_norm": 5.025066375732422,
      "learning_rate": 3.807274943668128e-05,
      "lm_loss": 0.006,
      "loss": 0.044,
      "step": 256
    },
    {
      "classification_loss": 0.0718,
      "epoch": 1.36158940397351,
      "grad_norm": 5.772749423980713,
      "learning_rate": 3.8046149838781846e-05,
      "lm_loss": 0.0097,
      "loss": 0.0815,
      "step": 257
    },
    {
      "classification_loss": 0.2125,
      "epoch": 1.366887417218543,
      "grad_norm": 25.51378059387207,
      "learning_rate": 3.801937735804838e-05,
      "lm_loss": 0.016,
      "loss": 0.2284,
      "step": 258
    },
    {
      "classification_loss": 0.7237,
      "epoch": 1.372185430463576,
      "grad_norm": 9.980001449584961,
      "learning_rate": 3.799243225096233e-05,
      "lm_loss": 0.0786,
      "loss": 0.8022,
      "step": 259
    },
    {
      "classification_loss": 0.1279,
      "epoch": 1.3774834437086092,
      "grad_norm": 4.669674396514893,
      "learning_rate": 3.7965314775658894e-05,
      "lm_loss": 0.0072,
      "loss": 0.1351,
      "step": 260
    },
    {
      "classification_loss": 0.1672,
      "epoch": 1.3827814569536425,
      "grad_norm": 6.550952911376953,
      "learning_rate": 3.7938025191924544e-05,
      "lm_loss": 0.018,
      "loss": 0.1852,
      "step": 261
    },
    {
      "classification_loss": 0.1458,
      "epoch": 1.3880794701986754,
      "grad_norm": 16.83785057067871,
      "learning_rate": 3.79105637611946e-05,
      "lm_loss": 0.0166,
      "loss": 0.1624,
      "step": 262
    },
    {
      "classification_loss": 0.3954,
      "epoch": 1.3933774834437087,
      "grad_norm": 12.447022438049316,
      "learning_rate": 3.788293074655066e-05,
      "lm_loss": 0.0314,
      "loss": 0.4268,
      "step": 263
    },
    {
      "classification_loss": 0.3522,
      "epoch": 1.3986754966887418,
      "grad_norm": 8.42427921295166,
      "learning_rate": 3.785512641271809e-05,
      "lm_loss": 0.0426,
      "loss": 0.3948,
      "step": 264
    },
    {
      "classification_loss": 0.1836,
      "epoch": 1.403973509933775,
      "grad_norm": 6.312130928039551,
      "learning_rate": 3.782715102606354e-05,
      "lm_loss": 0.0233,
      "loss": 0.2069,
      "step": 265
    },
    {
      "classification_loss": 0.1056,
      "epoch": 1.409271523178808,
      "grad_norm": 4.4287333488464355,
      "learning_rate": 3.77990048545923e-05,
      "lm_loss": 0.0126,
      "loss": 0.1182,
      "step": 266
    },
    {
      "classification_loss": 0.0444,
      "epoch": 1.414569536423841,
      "grad_norm": 3.9816722869873047,
      "learning_rate": 3.7770688167945825e-05,
      "lm_loss": 0.0061,
      "loss": 0.0506,
      "step": 267
    },
    {
      "classification_loss": 0.5143,
      "epoch": 1.4198675496688742,
      "grad_norm": 7.766134738922119,
      "learning_rate": 3.7742201237399105e-05,
      "lm_loss": 0.0482,
      "loss": 0.5625,
      "step": 268
    },
    {
      "classification_loss": 0.4201,
      "epoch": 1.4251655629139073,
      "grad_norm": 6.72153902053833,
      "learning_rate": 3.7713544335858075e-05,
      "lm_loss": 0.0375,
      "loss": 0.4576,
      "step": 269
    },
    {
      "classification_loss": 0.1121,
      "epoch": 1.4304635761589404,
      "grad_norm": 4.404346466064453,
      "learning_rate": 3.768471773785698e-05,
      "lm_loss": 0.0118,
      "loss": 0.124,
      "step": 270
    },
    {
      "classification_loss": 0.3182,
      "epoch": 1.4357615894039735,
      "grad_norm": 9.858338356018066,
      "learning_rate": 3.765572171955578e-05,
      "lm_loss": 0.0249,
      "loss": 0.3431,
      "step": 271
    },
    {
      "classification_loss": 0.6104,
      "epoch": 1.4410596026490066,
      "grad_norm": 16.76839256286621,
      "learning_rate": 3.762655655873748e-05,
      "lm_loss": 0.0431,
      "loss": 0.6536,
      "step": 272
    },
    {
      "classification_loss": 0.186,
      "epoch": 1.4463576158940397,
      "grad_norm": 25.645875930786133,
      "learning_rate": 3.759722253480549e-05,
      "lm_loss": 0.0253,
      "loss": 0.2112,
      "step": 273
    },
    {
      "classification_loss": 0.3057,
      "epoch": 1.451655629139073,
      "grad_norm": 6.152409553527832,
      "learning_rate": 3.756771992878093e-05,
      "lm_loss": 0.0378,
      "loss": 0.3435,
      "step": 274
    },
    {
      "classification_loss": 0.1621,
      "epoch": 1.4569536423841059,
      "grad_norm": 6.036374092102051,
      "learning_rate": 3.753804902329993e-05,
      "lm_loss": 0.015,
      "loss": 0.1771,
      "step": 275
    },
    {
      "classification_loss": 0.1189,
      "epoch": 1.4622516556291392,
      "grad_norm": 4.674655437469482,
      "learning_rate": 3.7508210102610955e-05,
      "lm_loss": 0.0099,
      "loss": 0.1288,
      "step": 276
    },
    {
      "classification_loss": 0.0589,
      "epoch": 1.4675496688741723,
      "grad_norm": 2.842445135116577,
      "learning_rate": 3.747820345257205e-05,
      "lm_loss": 0.0036,
      "loss": 0.0625,
      "step": 277
    },
    {
      "classification_loss": 0.3559,
      "epoch": 1.4728476821192054,
      "grad_norm": 8.19643497467041,
      "learning_rate": 3.744802936064812e-05,
      "lm_loss": 0.0372,
      "loss": 0.3931,
      "step": 278
    },
    {
      "classification_loss": 0.0428,
      "epoch": 1.4781456953642385,
      "grad_norm": 2.823199510574341,
      "learning_rate": 3.741768811590816e-05,
      "lm_loss": 0.0034,
      "loss": 0.0462,
      "step": 279
    },
    {
      "classification_loss": 0.3805,
      "epoch": 1.4834437086092715,
      "grad_norm": 7.799323558807373,
      "learning_rate": 3.73871800090225e-05,
      "lm_loss": 0.0275,
      "loss": 0.408,
      "step": 280
    },
    {
      "classification_loss": 0.0659,
      "epoch": 1.4887417218543046,
      "grad_norm": 3.3911259174346924,
      "learning_rate": 3.735650533226003e-05,
      "lm_loss": 0.0071,
      "loss": 0.073,
      "step": 281
    },
    {
      "classification_loss": 0.3655,
      "epoch": 1.4940397350993377,
      "grad_norm": 8.66239070892334,
      "learning_rate": 3.732566437948537e-05,
      "lm_loss": 0.038,
      "loss": 0.4035,
      "step": 282
    },
    {
      "classification_loss": 0.4028,
      "epoch": 1.4993377483443708,
      "grad_norm": 9.671128273010254,
      "learning_rate": 3.729465744615608e-05,
      "lm_loss": 0.0401,
      "loss": 0.4429,
      "step": 283
    },
    {
      "classification_loss": 0.334,
      "epoch": 1.504635761589404,
      "grad_norm": 9.753667831420898,
      "learning_rate": 3.7263484829319804e-05,
      "lm_loss": 0.039,
      "loss": 0.373,
      "step": 284
    },
    {
      "classification_loss": 0.0298,
      "epoch": 1.5099337748344372,
      "grad_norm": 1.506174087524414,
      "learning_rate": 3.723214682761147e-05,
      "lm_loss": 0.002,
      "loss": 0.0318,
      "step": 285
    },
    {
      "classification_loss": 0.0099,
      "epoch": 1.5152317880794701,
      "grad_norm": 0.4210752248764038,
      "learning_rate": 3.7200643741250365e-05,
      "lm_loss": 0.0014,
      "loss": 0.0113,
      "step": 286
    },
    {
      "classification_loss": 0.4927,
      "epoch": 1.5205298013245034,
      "grad_norm": 12.934358596801758,
      "learning_rate": 3.716897587203733e-05,
      "lm_loss": 0.0448,
      "loss": 0.5375,
      "step": 287
    },
    {
      "classification_loss": 0.5321,
      "epoch": 1.5258278145695363,
      "grad_norm": 19.386032104492188,
      "learning_rate": 3.7137143523351787e-05,
      "lm_loss": 0.0563,
      "loss": 0.5883,
      "step": 288
    },
    {
      "classification_loss": 0.227,
      "epoch": 1.5311258278145696,
      "grad_norm": 3.8577845096588135,
      "learning_rate": 3.710514700014892e-05,
      "lm_loss": 0.0241,
      "loss": 0.2511,
      "step": 289
    },
    {
      "classification_loss": 0.6296,
      "epoch": 1.5364238410596025,
      "grad_norm": 27.409175872802734,
      "learning_rate": 3.70729866089567e-05,
      "lm_loss": 0.0478,
      "loss": 0.6774,
      "step": 290
    },
    {
      "classification_loss": 0.2457,
      "epoch": 1.5417218543046358,
      "grad_norm": 7.641339302062988,
      "learning_rate": 3.704066265787295e-05,
      "lm_loss": 0.0248,
      "loss": 0.2705,
      "step": 291
    },
    {
      "classification_loss": 0.1934,
      "epoch": 1.547019867549669,
      "grad_norm": 15.186677932739258,
      "learning_rate": 3.70081754565624e-05,
      "lm_loss": 0.0178,
      "loss": 0.2112,
      "step": 292
    },
    {
      "classification_loss": 0.5636,
      "epoch": 1.552317880794702,
      "grad_norm": 35.84861373901367,
      "learning_rate": 3.6975525316253744e-05,
      "lm_loss": 0.0343,
      "loss": 0.5979,
      "step": 293
    },
    {
      "classification_loss": 0.3274,
      "epoch": 1.557615894039735,
      "grad_norm": 9.020503044128418,
      "learning_rate": 3.694271254973662e-05,
      "lm_loss": 0.033,
      "loss": 0.3605,
      "step": 294
    },
    {
      "classification_loss": 0.1899,
      "epoch": 1.5629139072847682,
      "grad_norm": 5.103623867034912,
      "learning_rate": 3.690973747135864e-05,
      "lm_loss": 0.0214,
      "loss": 0.2113,
      "step": 295
    },
    {
      "classification_loss": 0.3175,
      "epoch": 1.5682119205298013,
      "grad_norm": 8.844812393188477,
      "learning_rate": 3.6876600397022366e-05,
      "lm_loss": 0.0334,
      "loss": 0.351,
      "step": 296
    },
    {
      "classification_loss": 0.0957,
      "epoch": 1.5735099337748344,
      "grad_norm": 2.6427974700927734,
      "learning_rate": 3.6843301644182295e-05,
      "lm_loss": 0.0056,
      "loss": 0.1013,
      "step": 297
    },
    {
      "classification_loss": 0.2503,
      "epoch": 1.5788079470198677,
      "grad_norm": 4.6609978675842285,
      "learning_rate": 3.6809841531841796e-05,
      "lm_loss": 0.0229,
      "loss": 0.2732,
      "step": 298
    },
    {
      "classification_loss": 0.1513,
      "epoch": 1.5841059602649006,
      "grad_norm": 3.930695056915283,
      "learning_rate": 3.677622038055008e-05,
      "lm_loss": 0.0187,
      "loss": 0.17,
      "step": 299
    },
    {
      "classification_loss": 0.2227,
      "epoch": 1.589403973509934,
      "grad_norm": 6.081264972686768,
      "learning_rate": 3.6742438512399123e-05,
      "lm_loss": 0.0263,
      "loss": 0.249,
      "step": 300
    },
    {
      "epoch": 1.589403973509934,
      "eval_accuracy": 0.8845070600509644,
      "eval_auroc": 0.9503271579742432,
      "eval_classification_loss": 0.32174423336982727,
      "eval_f1": 0.8706624507904053,
      "eval_lm_loss": 0.03282451629638672,
      "eval_loss": 0.354568749666214,
      "eval_precision": 0.9583333134651184,
      "eval_recall": 0.7976878881454468,
      "eval_runtime": 60.1849,
      "eval_samples_per_second": 5.898,
      "eval_steps_per_second": 2.958,
      "step": 300
    },
    {
      "classification_loss": 0.2071,
      "epoch": 1.5947019867549668,
      "grad_norm": 9.868237495422363,
      "learning_rate": 3.6708496251020564e-05,
      "lm_loss": 0.0207,
      "loss": 0.2278,
      "step": 301
    },
    {
      "classification_loss": 0.0936,
      "epoch": 1.6,
      "grad_norm": 4.609295845031738,
      "learning_rate": 3.667439392158261e-05,
      "lm_loss": 0.013,
      "loss": 0.1066,
      "step": 302
    },
    {
      "classification_loss": 0.1538,
      "epoch": 1.605298013245033,
      "grad_norm": 5.210891246795654,
      "learning_rate": 3.6640131850786936e-05,
      "lm_loss": 0.0166,
      "loss": 0.1704,
      "step": 303
    },
    {
      "classification_loss": 0.1278,
      "epoch": 1.6105960264900663,
      "grad_norm": 10.839700698852539,
      "learning_rate": 3.660571036686556e-05,
      "lm_loss": 0.0141,
      "loss": 0.1419,
      "step": 304
    },
    {
      "classification_loss": 0.0152,
      "epoch": 1.6158940397350994,
      "grad_norm": 0.8716986179351807,
      "learning_rate": 3.6571129799577654e-05,
      "lm_loss": 0.0019,
      "loss": 0.0172,
      "step": 305
    },
    {
      "classification_loss": 0.0363,
      "epoch": 1.6211920529801325,
      "grad_norm": 3.5243327617645264,
      "learning_rate": 3.653639048020644e-05,
      "lm_loss": 0.0048,
      "loss": 0.0411,
      "step": 306
    },
    {
      "classification_loss": 0.0038,
      "epoch": 1.6264900662251656,
      "grad_norm": 0.2862269878387451,
      "learning_rate": 3.650149274155597e-05,
      "lm_loss": 0.0007,
      "loss": 0.0045,
      "step": 307
    },
    {
      "classification_loss": 0.4445,
      "epoch": 1.6317880794701987,
      "grad_norm": 18.574359893798828,
      "learning_rate": 3.646643691794797e-05,
      "lm_loss": 0.0304,
      "loss": 0.475,
      "step": 308
    },
    {
      "classification_loss": 0.0395,
      "epoch": 1.6370860927152318,
      "grad_norm": 12.024374961853027,
      "learning_rate": 3.643122334521863e-05,
      "lm_loss": 0.0058,
      "loss": 0.0453,
      "step": 309
    },
    {
      "classification_loss": 0.3202,
      "epoch": 1.6423841059602649,
      "grad_norm": 17.604482650756836,
      "learning_rate": 3.639585236071539e-05,
      "lm_loss": 0.0219,
      "loss": 0.3421,
      "step": 310
    },
    {
      "classification_loss": 0.7472,
      "epoch": 1.6476821192052982,
      "grad_norm": 9.913177490234375,
      "learning_rate": 3.636032430329367e-05,
      "lm_loss": 0.0533,
      "loss": 0.8006,
      "step": 311
    },
    {
      "classification_loss": 0.0003,
      "epoch": 1.652980132450331,
      "grad_norm": 0.031628869473934174,
      "learning_rate": 3.632463951331368e-05,
      "lm_loss": 0.0001,
      "loss": 0.0004,
      "step": 312
    },
    {
      "classification_loss": 0.0004,
      "epoch": 1.6582781456953644,
      "grad_norm": 0.12074816226959229,
      "learning_rate": 3.628879833263714e-05,
      "lm_loss": 0.0001,
      "loss": 0.0006,
      "step": 313
    },
    {
      "classification_loss": 0.0006,
      "epoch": 1.6635761589403972,
      "grad_norm": 0.043469030410051346,
      "learning_rate": 3.625280110462398e-05,
      "lm_loss": 0.0001,
      "loss": 0.0007,
      "step": 314
    },
    {
      "classification_loss": 0.4555,
      "epoch": 1.6688741721854305,
      "grad_norm": 8.978562355041504,
      "learning_rate": 3.621664817412908e-05,
      "lm_loss": 0.0673,
      "loss": 0.5228,
      "step": 315
    },
    {
      "classification_loss": 0.7496,
      "epoch": 1.6741721854304634,
      "grad_norm": 62.929786682128906,
      "learning_rate": 3.6180339887498953e-05,
      "lm_loss": 0.0815,
      "loss": 0.8311,
      "step": 316
    },
    {
      "classification_loss": 2.0981,
      "epoch": 1.6794701986754967,
      "grad_norm": 40.738853454589844,
      "learning_rate": 3.6143876592568426e-05,
      "lm_loss": 0.1762,
      "loss": 2.2743,
      "step": 317
    },
    {
      "classification_loss": 1.476,
      "epoch": 1.6847682119205298,
      "grad_norm": 22.212438583374023,
      "learning_rate": 3.6107258638657324e-05,
      "lm_loss": 0.1292,
      "loss": 1.6052,
      "step": 318
    },
    {
      "classification_loss": 0.8651,
      "epoch": 1.690066225165563,
      "grad_norm": 9.901700019836426,
      "learning_rate": 3.60704863765671e-05,
      "lm_loss": 0.0894,
      "loss": 0.9544,
      "step": 319
    },
    {
      "classification_loss": 0.7725,
      "epoch": 1.695364238410596,
      "grad_norm": 25.033493041992188,
      "learning_rate": 3.603356015857748e-05,
      "lm_loss": 0.0677,
      "loss": 0.8402,
      "step": 320
    },
    {
      "classification_loss": 0.0156,
      "epoch": 1.7006622516556291,
      "grad_norm": 0.7297423481941223,
      "learning_rate": 3.599648033844312e-05,
      "lm_loss": 0.0016,
      "loss": 0.0172,
      "step": 321
    },
    {
      "classification_loss": 0.4727,
      "epoch": 1.7059602649006622,
      "grad_norm": 6.520756244659424,
      "learning_rate": 3.5959247271390164e-05,
      "lm_loss": 0.0447,
      "loss": 0.5174,
      "step": 322
    },
    {
      "classification_loss": 0.6281,
      "epoch": 1.7112582781456953,
      "grad_norm": 35.83902359008789,
      "learning_rate": 3.592186131411288e-05,
      "lm_loss": 0.0471,
      "loss": 0.6752,
      "step": 323
    },
    {
      "classification_loss": 0.0265,
      "epoch": 1.7165562913907286,
      "grad_norm": 0.9867174029350281,
      "learning_rate": 3.5884322824770216e-05,
      "lm_loss": 0.0034,
      "loss": 0.03,
      "step": 324
    },
    {
      "classification_loss": 0.0249,
      "epoch": 1.7218543046357615,
      "grad_norm": 1.0298084020614624,
      "learning_rate": 3.584663216298242e-05,
      "lm_loss": 0.0034,
      "loss": 0.0283,
      "step": 325
    },
    {
      "classification_loss": 0.454,
      "epoch": 1.7271523178807948,
      "grad_norm": 14.185229301452637,
      "learning_rate": 3.58087896898275e-05,
      "lm_loss": 0.0553,
      "loss": 0.5093,
      "step": 326
    },
    {
      "classification_loss": 0.1202,
      "epoch": 1.7324503311258277,
      "grad_norm": 13.329483985900879,
      "learning_rate": 3.577079576783788e-05,
      "lm_loss": 0.0091,
      "loss": 0.1293,
      "step": 327
    },
    {
      "classification_loss": 0.1217,
      "epoch": 1.737748344370861,
      "grad_norm": 6.071735858917236,
      "learning_rate": 3.573265076099684e-05,
      "lm_loss": 0.0174,
      "loss": 0.1391,
      "step": 328
    },
    {
      "classification_loss": 0.0753,
      "epoch": 1.7430463576158939,
      "grad_norm": 9.520695686340332,
      "learning_rate": 3.569435503473505e-05,
      "lm_loss": 0.0078,
      "loss": 0.0832,
      "step": 329
    },
    {
      "classification_loss": 0.0145,
      "epoch": 1.7483443708609272,
      "grad_norm": 0.6467930674552917,
      "learning_rate": 3.565590895592712e-05,
      "lm_loss": 0.0012,
      "loss": 0.0157,
      "step": 330
    },
    {
      "classification_loss": 0.2747,
      "epoch": 1.7536423841059603,
      "grad_norm": 11.08447265625,
      "learning_rate": 3.561731289288799e-05,
      "lm_loss": 0.0259,
      "loss": 0.3007,
      "step": 331
    },
    {
      "classification_loss": 0.2641,
      "epoch": 1.7589403973509934,
      "grad_norm": 4.835735321044922,
      "learning_rate": 3.55785672153695e-05,
      "lm_loss": 0.0322,
      "loss": 0.2963,
      "step": 332
    },
    {
      "classification_loss": 0.3303,
      "epoch": 1.7642384105960265,
      "grad_norm": 10.318045616149902,
      "learning_rate": 3.553967229455676e-05,
      "lm_loss": 0.0377,
      "loss": 0.368,
      "step": 333
    },
    {
      "classification_loss": 0.2162,
      "epoch": 1.7695364238410596,
      "grad_norm": 9.991059303283691,
      "learning_rate": 3.5500628503064685e-05,
      "lm_loss": 0.0317,
      "loss": 0.2479,
      "step": 334
    },
    {
      "classification_loss": 0.3982,
      "epoch": 1.7748344370860927,
      "grad_norm": 14.768351554870605,
      "learning_rate": 3.5461436214934344e-05,
      "lm_loss": 0.0215,
      "loss": 0.4197,
      "step": 335
    },
    {
      "classification_loss": 0.1244,
      "epoch": 1.7801324503311258,
      "grad_norm": 4.412431716918945,
      "learning_rate": 3.5422095805629415e-05,
      "lm_loss": 0.0187,
      "loss": 0.1431,
      "step": 336
    },
    {
      "classification_loss": 0.0828,
      "epoch": 1.785430463576159,
      "grad_norm": 3.7633635997772217,
      "learning_rate": 3.538260765203258e-05,
      "lm_loss": 0.0038,
      "loss": 0.0867,
      "step": 337
    },
    {
      "classification_loss": 0.3114,
      "epoch": 1.790728476821192,
      "grad_norm": 8.21419620513916,
      "learning_rate": 3.534297213244193e-05,
      "lm_loss": 0.0302,
      "loss": 0.3416,
      "step": 338
    },
    {
      "classification_loss": 0.4337,
      "epoch": 1.7960264900662253,
      "grad_norm": 7.491161346435547,
      "learning_rate": 3.530318962656733e-05,
      "lm_loss": 0.0332,
      "loss": 0.4669,
      "step": 339
    },
    {
      "classification_loss": 0.1344,
      "epoch": 1.8013245033112582,
      "grad_norm": 4.456019878387451,
      "learning_rate": 3.526326051552675e-05,
      "lm_loss": 0.0181,
      "loss": 0.1525,
      "step": 340
    },
    {
      "classification_loss": 0.4687,
      "epoch": 1.8066225165562915,
      "grad_norm": 14.532021522521973,
      "learning_rate": 3.522318518184268e-05,
      "lm_loss": 0.0394,
      "loss": 0.5081,
      "step": 341
    },
    {
      "classification_loss": 0.3895,
      "epoch": 1.8119205298013243,
      "grad_norm": 10.936323165893555,
      "learning_rate": 3.51829640094384e-05,
      "lm_loss": 0.0381,
      "loss": 0.4275,
      "step": 342
    },
    {
      "classification_loss": 0.1186,
      "epoch": 1.8172185430463577,
      "grad_norm": 4.368919849395752,
      "learning_rate": 3.514259738363436e-05,
      "lm_loss": 0.0133,
      "loss": 0.1318,
      "step": 343
    },
    {
      "classification_loss": 0.2576,
      "epoch": 1.8225165562913908,
      "grad_norm": 7.333866596221924,
      "learning_rate": 3.510208569114442e-05,
      "lm_loss": 0.0231,
      "loss": 0.2807,
      "step": 344
    },
    {
      "classification_loss": 0.1769,
      "epoch": 1.8278145695364238,
      "grad_norm": 4.656265735626221,
      "learning_rate": 3.5061429320072225e-05,
      "lm_loss": 0.0243,
      "loss": 0.2012,
      "step": 345
    },
    {
      "classification_loss": 0.1094,
      "epoch": 1.833112582781457,
      "grad_norm": 3.854640007019043,
      "learning_rate": 3.502062865990742e-05,
      "lm_loss": 0.0106,
      "loss": 0.1199,
      "step": 346
    },
    {
      "classification_loss": 0.083,
      "epoch": 1.83841059602649,
      "grad_norm": 3.1809322834014893,
      "learning_rate": 3.497968410152196e-05,
      "lm_loss": 0.0128,
      "loss": 0.0958,
      "step": 347
    },
    {
      "classification_loss": 0.2893,
      "epoch": 1.8437086092715231,
      "grad_norm": 15.538369178771973,
      "learning_rate": 3.493859603716634e-05,
      "lm_loss": 0.0311,
      "loss": 0.3204,
      "step": 348
    },
    {
      "classification_loss": 0.1252,
      "epoch": 1.8490066225165562,
      "grad_norm": 7.439122200012207,
      "learning_rate": 3.489736486046585e-05,
      "lm_loss": 0.0101,
      "loss": 0.1353,
      "step": 349
    },
    {
      "classification_loss": 0.0245,
      "epoch": 1.8543046357615895,
      "grad_norm": 2.0325450897216797,
      "learning_rate": 3.485599096641682e-05,
      "lm_loss": 0.0051,
      "loss": 0.0295,
      "step": 350
    },
    {
      "epoch": 1.8543046357615895,
      "eval_accuracy": 0.8845070600509644,
      "eval_auroc": 0.9502954483032227,
      "eval_classification_loss": 0.34444305300712585,
      "eval_f1": 0.8851540684700012,
      "eval_lm_loss": 0.031157689169049263,
      "eval_loss": 0.3756007254123688,
      "eval_precision": 0.8586956262588501,
      "eval_recall": 0.913294792175293,
      "eval_runtime": 59.8093,
      "eval_samples_per_second": 5.936,
      "eval_steps_per_second": 2.976,
      "step": 350
    },
    {
      "classification_loss": 0.0472,
      "epoch": 1.8596026490066224,
      "grad_norm": 2.4528257846832275,
      "learning_rate": 3.481447475138279e-05,
      "lm_loss": 0.0083,
      "loss": 0.0555,
      "step": 351
    },
    {
      "classification_loss": 0.0067,
      "epoch": 1.8649006622516557,
      "grad_norm": 0.3057335317134857,
      "learning_rate": 3.4772816613090745e-05,
      "lm_loss": 0.0015,
      "loss": 0.0082,
      "step": 352
    },
    {
      "classification_loss": 0.3526,
      "epoch": 1.8701986754966886,
      "grad_norm": 7.811878204345703,
      "learning_rate": 3.473101695062732e-05,
      "lm_loss": 0.0454,
      "loss": 0.398,
      "step": 353
    },
    {
      "classification_loss": 1.2411,
      "epoch": 1.875496688741722,
      "grad_norm": 21.15567398071289,
      "learning_rate": 3.4689076164434954e-05,
      "lm_loss": 0.0904,
      "loss": 1.3315,
      "step": 354
    },
    {
      "classification_loss": 0.3973,
      "epoch": 1.8807947019867548,
      "grad_norm": 6.591670036315918,
      "learning_rate": 3.464699465630803e-05,
      "lm_loss": 0.0455,
      "loss": 0.4428,
      "step": 355
    },
    {
      "classification_loss": 0.3823,
      "epoch": 1.8860927152317881,
      "grad_norm": 18.370065689086914,
      "learning_rate": 3.460477282938908e-05,
      "lm_loss": 0.0354,
      "loss": 0.4177,
      "step": 356
    },
    {
      "classification_loss": 0.0885,
      "epoch": 1.8913907284768212,
      "grad_norm": 13.339816093444824,
      "learning_rate": 3.456241108816489e-05,
      "lm_loss": 0.0096,
      "loss": 0.0981,
      "step": 357
    },
    {
      "classification_loss": 0.0068,
      "epoch": 1.8966887417218543,
      "grad_norm": 1.0805500745773315,
      "learning_rate": 3.451990983846262e-05,
      "lm_loss": 0.0019,
      "loss": 0.0086,
      "step": 358
    },
    {
      "classification_loss": 0.5724,
      "epoch": 1.9019867549668874,
      "grad_norm": 18.253080368041992,
      "learning_rate": 3.4477269487445944e-05,
      "lm_loss": 0.0445,
      "loss": 0.6169,
      "step": 359
    },
    {
      "classification_loss": 0.7135,
      "epoch": 1.9072847682119205,
      "grad_norm": 18.70496940612793,
      "learning_rate": 3.443449044361112e-05,
      "lm_loss": 0.0708,
      "loss": 0.7843,
      "step": 360
    },
    {
      "classification_loss": 0.0053,
      "epoch": 1.9125827814569536,
      "grad_norm": 0.26829659938812256,
      "learning_rate": 3.439157311678308e-05,
      "lm_loss": 0.001,
      "loss": 0.0063,
      "step": 361
    },
    {
      "classification_loss": 0.4688,
      "epoch": 1.9178807947019867,
      "grad_norm": 21.069429397583008,
      "learning_rate": 3.434851791811154e-05,
      "lm_loss": 0.0473,
      "loss": 0.5161,
      "step": 362
    },
    {
      "classification_loss": 0.3957,
      "epoch": 1.92317880794702,
      "grad_norm": 15.419574737548828,
      "learning_rate": 3.430532526006701e-05,
      "lm_loss": 0.0313,
      "loss": 0.427,
      "step": 363
    },
    {
      "classification_loss": 0.4972,
      "epoch": 1.9284768211920529,
      "grad_norm": 12.177935600280762,
      "learning_rate": 3.426199555643687e-05,
      "lm_loss": 0.0509,
      "loss": 0.5481,
      "step": 364
    },
    {
      "classification_loss": 0.0214,
      "epoch": 1.9337748344370862,
      "grad_norm": 2.425583600997925,
      "learning_rate": 3.421852922232142e-05,
      "lm_loss": 0.0021,
      "loss": 0.0235,
      "step": 365
    },
    {
      "classification_loss": 0.0819,
      "epoch": 1.939072847682119,
      "grad_norm": 8.522496223449707,
      "learning_rate": 3.4174926674129855e-05,
      "lm_loss": 0.011,
      "loss": 0.0929,
      "step": 366
    },
    {
      "classification_loss": 0.2696,
      "epoch": 1.9443708609271524,
      "grad_norm": 11.291783332824707,
      "learning_rate": 3.413118832957633e-05,
      "lm_loss": 0.0188,
      "loss": 0.2884,
      "step": 367
    },
    {
      "classification_loss": 0.0775,
      "epoch": 1.9496688741721855,
      "grad_norm": 7.856764316558838,
      "learning_rate": 3.408731460767593e-05,
      "lm_loss": 0.0087,
      "loss": 0.0862,
      "step": 368
    },
    {
      "classification_loss": 0.271,
      "epoch": 1.9549668874172186,
      "grad_norm": 7.723100662231445,
      "learning_rate": 3.404330592874065e-05,
      "lm_loss": 0.0261,
      "loss": 0.297,
      "step": 369
    },
    {
      "classification_loss": 0.4038,
      "epoch": 1.9602649006622517,
      "grad_norm": 6.125762462615967,
      "learning_rate": 3.3999162714375405e-05,
      "lm_loss": 0.0417,
      "loss": 0.4455,
      "step": 370
    },
    {
      "classification_loss": 0.0703,
      "epoch": 1.9655629139072848,
      "grad_norm": 5.4816741943359375,
      "learning_rate": 3.395488538747393e-05,
      "lm_loss": 0.006,
      "loss": 0.0763,
      "step": 371
    },
    {
      "classification_loss": 0.0567,
      "epoch": 1.9708609271523179,
      "grad_norm": 3.1961610317230225,
      "learning_rate": 3.3910474372214773e-05,
      "lm_loss": 0.0041,
      "loss": 0.0608,
      "step": 372
    },
    {
      "classification_loss": 0.5009,
      "epoch": 1.976158940397351,
      "grad_norm": 12.27746295928955,
      "learning_rate": 3.386593009405724e-05,
      "lm_loss": 0.0551,
      "loss": 0.556,
      "step": 373
    },
    {
      "classification_loss": 0.2248,
      "epoch": 1.981456953642384,
      "grad_norm": 6.058496952056885,
      "learning_rate": 3.38212529797373e-05,
      "lm_loss": 0.02,
      "loss": 0.2448,
      "step": 374
    },
    {
      "classification_loss": 0.0322,
      "epoch": 1.9867549668874172,
      "grad_norm": 1.3348891735076904,
      "learning_rate": 3.377644345726347e-05,
      "lm_loss": 0.0027,
      "loss": 0.0349,
      "step": 375
    },
    {
      "classification_loss": 0.0556,
      "epoch": 1.9920529801324505,
      "grad_norm": 3.9021286964416504,
      "learning_rate": 3.3731501955912754e-05,
      "lm_loss": 0.0056,
      "loss": 0.0611,
      "step": 376
    },
    {
      "classification_loss": 0.0653,
      "epoch": 1.9973509933774833,
      "grad_norm": 3.6003711223602295,
      "learning_rate": 3.3686428906226544e-05,
      "lm_loss": 0.0049,
      "loss": 0.0701,
      "step": 377
    },
    {
      "classification_loss": 0.0194,
      "epoch": 2.0026490066225167,
      "grad_norm": 0.7261447310447693,
      "learning_rate": 3.364122474000644e-05,
      "lm_loss": 0.0011,
      "loss": 0.0205,
      "step": 378
    },
    {
      "classification_loss": 0.0162,
      "epoch": 2.0079470198675495,
      "grad_norm": 0.6416781544685364,
      "learning_rate": 3.359588989031016e-05,
      "lm_loss": 0.0013,
      "loss": 0.0174,
      "step": 379
    },
    {
      "classification_loss": 0.0149,
      "epoch": 2.013245033112583,
      "grad_norm": 1.001346230506897,
      "learning_rate": 3.355042479144737e-05,
      "lm_loss": 0.0012,
      "loss": 0.0161,
      "step": 380
    },
    {
      "classification_loss": 0.0051,
      "epoch": 2.0185430463576157,
      "grad_norm": 0.19863009452819824,
      "learning_rate": 3.350482987897551e-05,
      "lm_loss": 0.0004,
      "loss": 0.0055,
      "step": 381
    },
    {
      "classification_loss": 0.0063,
      "epoch": 2.023841059602649,
      "grad_norm": 0.5387454628944397,
      "learning_rate": 3.345910558969568e-05,
      "lm_loss": 0.0009,
      "loss": 0.0072,
      "step": 382
    },
    {
      "classification_loss": 0.0023,
      "epoch": 2.029139072847682,
      "grad_norm": 0.0871424749493599,
      "learning_rate": 3.341325236164837e-05,
      "lm_loss": 0.0005,
      "loss": 0.0028,
      "step": 383
    },
    {
      "classification_loss": 0.0788,
      "epoch": 2.0344370860927152,
      "grad_norm": 10.270561218261719,
      "learning_rate": 3.3367270634109355e-05,
      "lm_loss": 0.0065,
      "loss": 0.0852,
      "step": 384
    },
    {
      "classification_loss": 0.0013,
      "epoch": 2.0397350993377485,
      "grad_norm": 0.05430527403950691,
      "learning_rate": 3.3321160847585395e-05,
      "lm_loss": 0.0003,
      "loss": 0.0015,
      "step": 385
    },
    {
      "classification_loss": 0.0009,
      "epoch": 2.0450331125827814,
      "grad_norm": 0.047437142580747604,
      "learning_rate": 3.3274923443810094e-05,
      "lm_loss": 0.0003,
      "loss": 0.0012,
      "step": 386
    },
    {
      "classification_loss": 0.0041,
      "epoch": 2.0503311258278147,
      "grad_norm": 1.4194906949996948,
      "learning_rate": 3.322855886573961e-05,
      "lm_loss": 0.0011,
      "loss": 0.0052,
      "step": 387
    },
    {
      "classification_loss": 0.0004,
      "epoch": 2.0556291390728476,
      "grad_norm": 0.02894170954823494,
      "learning_rate": 3.318206755754846e-05,
      "lm_loss": 0.0003,
      "loss": 0.0008,
      "step": 388
    },
    {
      "classification_loss": 0.2943,
      "epoch": 2.060927152317881,
      "grad_norm": 41.563720703125,
      "learning_rate": 3.313544996462522e-05,
      "lm_loss": 0.0202,
      "loss": 0.3145,
      "step": 389
    },
    {
      "classification_loss": 0.4642,
      "epoch": 2.066225165562914,
      "grad_norm": 6.95037317276001,
      "learning_rate": 3.308870653356828e-05,
      "lm_loss": 0.0444,
      "loss": 0.5087,
      "step": 390
    },
    {
      "classification_loss": 0.0003,
      "epoch": 2.071523178807947,
      "grad_norm": 0.020254751667380333,
      "learning_rate": 3.304183771218159e-05,
      "lm_loss": 0.0002,
      "loss": 0.0005,
      "step": 391
    },
    {
      "classification_loss": 0.0003,
      "epoch": 2.07682119205298,
      "grad_norm": 0.030997836962342262,
      "learning_rate": 3.299484394947031e-05,
      "lm_loss": 0.0002,
      "loss": 0.0005,
      "step": 392
    },
    {
      "classification_loss": 1.0014,
      "epoch": 2.0821192052980133,
      "grad_norm": 22.05500602722168,
      "learning_rate": 3.294772569563656e-05,
      "lm_loss": 0.0877,
      "loss": 1.0891,
      "step": 393
    },
    {
      "classification_loss": 0.3045,
      "epoch": 2.087417218543046,
      "grad_norm": 30.641328811645508,
      "learning_rate": 3.290048340207508e-05,
      "lm_loss": 0.0255,
      "loss": 0.3299,
      "step": 394
    },
    {
      "classification_loss": 0.016,
      "epoch": 2.0927152317880795,
      "grad_norm": 19.87793731689453,
      "learning_rate": 3.2853117521368944e-05,
      "lm_loss": 0.002,
      "loss": 0.0181,
      "step": 395
    },
    {
      "classification_loss": 0.0004,
      "epoch": 2.0980132450331124,
      "grad_norm": 0.03624943643808365,
      "learning_rate": 3.280562850728516e-05,
      "lm_loss": 0.0002,
      "loss": 0.0006,
      "step": 396
    },
    {
      "classification_loss": 0.0005,
      "epoch": 2.1033112582781457,
      "grad_norm": 0.02493858151137829,
      "learning_rate": 3.275801681477038e-05,
      "lm_loss": 0.0002,
      "loss": 0.0007,
      "step": 397
    },
    {
      "classification_loss": 0.0005,
      "epoch": 2.108609271523179,
      "grad_norm": 0.026823678985238075,
      "learning_rate": 3.271028289994651e-05,
      "lm_loss": 0.0002,
      "loss": 0.0007,
      "step": 398
    },
    {
      "classification_loss": 0.2453,
      "epoch": 2.113907284768212,
      "grad_norm": 61.75249481201172,
      "learning_rate": 3.266242722010636e-05,
      "lm_loss": 0.0158,
      "loss": 0.2611,
      "step": 399
    },
    {
      "classification_loss": 0.7784,
      "epoch": 2.119205298013245,
      "grad_norm": 72.33714294433594,
      "learning_rate": 3.2614450233709245e-05,
      "lm_loss": 0.0483,
      "loss": 0.8267,
      "step": 400
    },
    {
      "epoch": 2.119205298013245,
      "eval_accuracy": 0.8873239159584045,
      "eval_auroc": 0.9473575353622437,
      "eval_classification_loss": 0.6966193318367004,
      "eval_f1": 0.8857142925262451,
      "eval_lm_loss": 0.05669226124882698,
      "eval_loss": 0.7533116340637207,
      "eval_precision": 0.8757061958312988,
      "eval_recall": 0.8959537744522095,
      "eval_runtime": 61.6963,
      "eval_samples_per_second": 5.754,
      "eval_steps_per_second": 2.885,
      "step": 400
    },
    {
      "classification_loss": 0.001,
      "epoch": 2.124503311258278,
      "grad_norm": 0.05150590091943741,
      "learning_rate": 3.256635240037659e-05,
      "lm_loss": 0.0002,
      "loss": 0.0012,
      "step": 401
    },
    {
      "classification_loss": 0.0007,
      "epoch": 2.1298013245033114,
      "grad_norm": 0.0385078527033329,
      "learning_rate": 3.251813418088759e-05,
      "lm_loss": 0.0002,
      "loss": 0.0009,
      "step": 402
    },
    {
      "classification_loss": 0.1677,
      "epoch": 2.1350993377483443,
      "grad_norm": 36.44617462158203,
      "learning_rate": 3.246979603717467e-05,
      "lm_loss": 0.0129,
      "loss": 0.1806,
      "step": 403
    },
    {
      "classification_loss": 0.0007,
      "epoch": 2.1403973509933776,
      "grad_norm": 0.5752580165863037,
      "learning_rate": 3.24213384323192e-05,
      "lm_loss": 0.0006,
      "loss": 0.0014,
      "step": 404
    },
    {
      "classification_loss": 0.0008,
      "epoch": 2.1456953642384105,
      "grad_norm": 0.03988804295659065,
      "learning_rate": 3.2372761830546945e-05,
      "lm_loss": 0.0002,
      "loss": 0.0009,
      "step": 405
    },
    {
      "classification_loss": 0.001,
      "epoch": 2.1509933774834438,
      "grad_norm": 0.08262227475643158,
      "learning_rate": 3.232406669722369e-05,
      "lm_loss": 0.0003,
      "loss": 0.0012,
      "step": 406
    },
    {
      "classification_loss": 0.3988,
      "epoch": 2.1562913907284766,
      "grad_norm": 4.35977840423584,
      "learning_rate": 3.2275253498850754e-05,
      "lm_loss": 0.035,
      "loss": 0.4338,
      "step": 407
    },
    {
      "classification_loss": 0.0007,
      "epoch": 2.16158940397351,
      "grad_norm": 0.03520366549491882,
      "learning_rate": 3.222632270306052e-05,
      "lm_loss": 0.0002,
      "loss": 0.0009,
      "step": 408
    },
    {
      "classification_loss": 0.1512,
      "epoch": 2.1668874172185433,
      "grad_norm": 22.84488296508789,
      "learning_rate": 3.217727477861196e-05,
      "lm_loss": 0.012,
      "loss": 0.1632,
      "step": 409
    },
    {
      "classification_loss": 0.1584,
      "epoch": 2.172185430463576,
      "grad_norm": 77.10163879394531,
      "learning_rate": 3.212811019538614e-05,
      "lm_loss": 0.0102,
      "loss": 0.1686,
      "step": 410
    },
    {
      "classification_loss": 0.0025,
      "epoch": 2.1774834437086095,
      "grad_norm": 0.21661698818206787,
      "learning_rate": 3.207882942438173e-05,
      "lm_loss": 0.0006,
      "loss": 0.0031,
      "step": 411
    },
    {
      "classification_loss": 0.0016,
      "epoch": 2.1827814569536423,
      "grad_norm": 0.07945861667394638,
      "learning_rate": 3.202943293771047e-05,
      "lm_loss": 0.0003,
      "loss": 0.0019,
      "step": 412
    },
    {
      "classification_loss": 0.0019,
      "epoch": 2.1880794701986757,
      "grad_norm": 0.0921841487288475,
      "learning_rate": 3.197992120859268e-05,
      "lm_loss": 0.0003,
      "loss": 0.0022,
      "step": 413
    },
    {
      "classification_loss": 0.0228,
      "epoch": 2.1933774834437085,
      "grad_norm": 8.117704391479492,
      "learning_rate": 3.193029471135269e-05,
      "lm_loss": 0.0051,
      "loss": 0.0279,
      "step": 414
    },
    {
      "classification_loss": 0.8893,
      "epoch": 2.198675496688742,
      "grad_norm": 13.654373168945312,
      "learning_rate": 3.188055392141432e-05,
      "lm_loss": 0.0786,
      "loss": 0.9678,
      "step": 415
    },
    {
      "classification_loss": 0.2646,
      "epoch": 2.2039735099337747,
      "grad_norm": 39.09990310668945,
      "learning_rate": 3.183069931529632e-05,
      "lm_loss": 0.0137,
      "loss": 0.2783,
      "step": 416
    },
    {
      "classification_loss": 0.7798,
      "epoch": 2.209271523178808,
      "grad_norm": 7.6568522453308105,
      "learning_rate": 3.1780731370607804e-05,
      "lm_loss": 0.0558,
      "loss": 0.8356,
      "step": 417
    },
    {
      "classification_loss": 0.0043,
      "epoch": 2.214569536423841,
      "grad_norm": 0.2868093252182007,
      "learning_rate": 3.173065056604366e-05,
      "lm_loss": 0.0007,
      "loss": 0.005,
      "step": 418
    },
    {
      "classification_loss": 0.0084,
      "epoch": 2.2198675496688742,
      "grad_norm": 1.371382474899292,
      "learning_rate": 3.168045738138001e-05,
      "lm_loss": 0.0016,
      "loss": 0.0101,
      "step": 419
    },
    {
      "classification_loss": 0.0029,
      "epoch": 2.225165562913907,
      "grad_norm": 0.13364127278327942,
      "learning_rate": 3.1630152297469544e-05,
      "lm_loss": 0.0003,
      "loss": 0.0033,
      "step": 420
    },
    {
      "classification_loss": 0.0025,
      "epoch": 2.2304635761589404,
      "grad_norm": 0.11783082783222198,
      "learning_rate": 3.157973579623698e-05,
      "lm_loss": 0.0004,
      "loss": 0.0029,
      "step": 421
    },
    {
      "classification_loss": 0.002,
      "epoch": 2.2357615894039737,
      "grad_norm": 0.0880235806107521,
      "learning_rate": 3.152920836067439e-05,
      "lm_loss": 0.0003,
      "loss": 0.0023,
      "step": 422
    },
    {
      "classification_loss": 0.0015,
      "epoch": 2.2410596026490066,
      "grad_norm": 0.06724422425031662,
      "learning_rate": 3.1478570474836646e-05,
      "lm_loss": 0.0002,
      "loss": 0.0017,
      "step": 423
    },
    {
      "classification_loss": 0.0008,
      "epoch": 2.24635761589404,
      "grad_norm": 0.036194466054439545,
      "learning_rate": 3.142782262383669e-05,
      "lm_loss": 0.0002,
      "loss": 0.001,
      "step": 424
    },
    {
      "classification_loss": 0.3106,
      "epoch": 2.251655629139073,
      "grad_norm": 12.94101619720459,
      "learning_rate": 3.137696529384096e-05,
      "lm_loss": 0.0314,
      "loss": 0.342,
      "step": 425
    },
    {
      "classification_loss": 0.0142,
      "epoch": 2.256953642384106,
      "grad_norm": 7.555570125579834,
      "learning_rate": 3.1325998972064706e-05,
      "lm_loss": 0.0015,
      "loss": 0.0157,
      "step": 426
    },
    {
      "classification_loss": 0.0105,
      "epoch": 2.262251655629139,
      "grad_norm": 1.6550226211547852,
      "learning_rate": 3.1274924146767316e-05,
      "lm_loss": 0.0016,
      "loss": 0.0121,
      "step": 427
    },
    {
      "classification_loss": 0.3725,
      "epoch": 2.2675496688741723,
      "grad_norm": 8.158663749694824,
      "learning_rate": 3.122374130724765e-05,
      "lm_loss": 0.0299,
      "loss": 0.4024,
      "step": 428
    },
    {
      "classification_loss": 0.4593,
      "epoch": 2.272847682119205,
      "grad_norm": 4.891383171081543,
      "learning_rate": 3.117245094383935e-05,
      "lm_loss": 0.0471,
      "loss": 0.5065,
      "step": 429
    },
    {
      "classification_loss": 0.0114,
      "epoch": 2.2781456953642385,
      "grad_norm": 2.424532175064087,
      "learning_rate": 3.112105354790615e-05,
      "lm_loss": 0.0021,
      "loss": 0.0135,
      "step": 430
    },
    {
      "classification_loss": 0.0027,
      "epoch": 2.2834437086092714,
      "grad_norm": 0.12365718185901642,
      "learning_rate": 3.1069549611837124e-05,
      "lm_loss": 0.0003,
      "loss": 0.0031,
      "step": 431
    },
    {
      "classification_loss": 0.0018,
      "epoch": 2.2887417218543047,
      "grad_norm": 0.08102871477603912,
      "learning_rate": 3.101793962904205e-05,
      "lm_loss": 0.0003,
      "loss": 0.002,
      "step": 432
    },
    {
      "classification_loss": 0.0022,
      "epoch": 2.2940397350993376,
      "grad_norm": 0.12729793787002563,
      "learning_rate": 3.096622409394662e-05,
      "lm_loss": 0.0003,
      "loss": 0.0025,
      "step": 433
    },
    {
      "classification_loss": 0.0024,
      "epoch": 2.299337748344371,
      "grad_norm": 0.10251695662736893,
      "learning_rate": 3.091440350198771e-05,
      "lm_loss": 0.0005,
      "loss": 0.0029,
      "step": 434
    },
    {
      "classification_loss": 0.003,
      "epoch": 2.304635761589404,
      "grad_norm": 0.13450250029563904,
      "learning_rate": 3.086247834960866e-05,
      "lm_loss": 0.0004,
      "loss": 0.0035,
      "step": 435
    },
    {
      "classification_loss": 0.007,
      "epoch": 2.309933774834437,
      "grad_norm": 3.7453274726867676,
      "learning_rate": 3.08104491342545e-05,
      "lm_loss": 0.0016,
      "loss": 0.0086,
      "step": 436
    },
    {
      "classification_loss": 0.0301,
      "epoch": 2.3152317880794704,
      "grad_norm": 15.741969108581543,
      "learning_rate": 3.075831635436717e-05,
      "lm_loss": 0.0036,
      "loss": 0.0337,
      "step": 437
    },
    {
      "classification_loss": 0.1637,
      "epoch": 2.3205298013245033,
      "grad_norm": 20.153507232666016,
      "learning_rate": 3.070608050938079e-05,
      "lm_loss": 0.009,
      "loss": 0.1728,
      "step": 438
    },
    {
      "classification_loss": 0.0041,
      "epoch": 2.3258278145695366,
      "grad_norm": 0.6864255666732788,
      "learning_rate": 3.065374209971683e-05,
      "lm_loss": 0.0009,
      "loss": 0.0049,
      "step": 439
    },
    {
      "classification_loss": 0.0029,
      "epoch": 2.3311258278145695,
      "grad_norm": 0.16915327310562134,
      "learning_rate": 3.060130162677935e-05,
      "lm_loss": 0.0005,
      "loss": 0.0035,
      "step": 440
    },
    {
      "classification_loss": 0.0964,
      "epoch": 2.3364238410596028,
      "grad_norm": 52.540287017822266,
      "learning_rate": 3.0548759592950144e-05,
      "lm_loss": 0.009,
      "loss": 0.1054,
      "step": 441
    },
    {
      "classification_loss": 0.0038,
      "epoch": 2.3417218543046356,
      "grad_norm": 0.1648062914609909,
      "learning_rate": 3.0496116501584015e-05,
      "lm_loss": 0.0006,
      "loss": 0.0043,
      "step": 442
    },
    {
      "classification_loss": 0.0671,
      "epoch": 2.347019867549669,
      "grad_norm": 7.819965839385986,
      "learning_rate": 3.0443372857003857e-05,
      "lm_loss": 0.0056,
      "loss": 0.0727,
      "step": 443
    },
    {
      "classification_loss": 0.0009,
      "epoch": 2.352317880794702,
      "grad_norm": 0.03931473195552826,
      "learning_rate": 3.0390529164495898e-05,
      "lm_loss": 0.0003,
      "loss": 0.0012,
      "step": 444
    },
    {
      "classification_loss": 0.0018,
      "epoch": 2.357615894039735,
      "grad_norm": 0.08275849372148514,
      "learning_rate": 3.033758593030481e-05,
      "lm_loss": 0.0004,
      "loss": 0.0022,
      "step": 445
    },
    {
      "classification_loss": 0.0048,
      "epoch": 2.362913907284768,
      "grad_norm": 3.298156261444092,
      "learning_rate": 3.028454366162888e-05,
      "lm_loss": 0.0009,
      "loss": 0.0057,
      "step": 446
    },
    {
      "classification_loss": 0.001,
      "epoch": 2.3682119205298013,
      "grad_norm": 0.043255001306533813,
      "learning_rate": 3.0231402866615162e-05,
      "lm_loss": 0.0003,
      "loss": 0.0013,
      "step": 447
    },
    {
      "classification_loss": 0.0009,
      "epoch": 2.3735099337748347,
      "grad_norm": 0.0353359654545784,
      "learning_rate": 3.0178164054354582e-05,
      "lm_loss": 0.0001,
      "loss": 0.001,
      "step": 448
    },
    {
      "classification_loss": 0.0013,
      "epoch": 2.3788079470198675,
      "grad_norm": 0.05173790454864502,
      "learning_rate": 3.012482773487708e-05,
      "lm_loss": 0.0002,
      "loss": 0.0014,
      "step": 449
    },
    {
      "classification_loss": 0.0014,
      "epoch": 2.384105960264901,
      "grad_norm": 0.05724985897541046,
      "learning_rate": 3.0071394419146734e-05,
      "lm_loss": 0.0002,
      "loss": 0.0016,
      "step": 450
    },
    {
      "epoch": 2.384105960264901,
      "eval_accuracy": 0.9154929518699646,
      "eval_auroc": 0.9610620737075806,
      "eval_classification_loss": 0.4825713336467743,
      "eval_f1": 0.9096385836601257,
      "eval_lm_loss": 0.044623661786317825,
      "eval_loss": 0.5271949172019958,
      "eval_precision": 0.9496855139732361,
      "eval_recall": 0.8728323578834534,
      "eval_runtime": 62.292,
      "eval_samples_per_second": 5.699,
      "eval_steps_per_second": 2.858,
      "step": 450
    },
    {
      "classification_loss": 0.0018,
      "epoch": 2.3894039735099337,
      "grad_norm": 0.07709845900535583,
      "learning_rate": 3.001786461905682e-05,
      "lm_loss": 0.0001,
      "loss": 0.0019,
      "step": 451
    },
    {
      "classification_loss": 0.0015,
      "epoch": 2.394701986754967,
      "grad_norm": 0.11981970071792603,
      "learning_rate": 2.996423884742496e-05,
      "lm_loss": 0.0003,
      "loss": 0.0018,
      "step": 452
    },
    {
      "classification_loss": 0.0048,
      "epoch": 2.4,
      "grad_norm": 1.603007435798645,
      "learning_rate": 2.9910517617988192e-05,
      "lm_loss": 0.0014,
      "loss": 0.0062,
      "step": 453
    },
    {
      "classification_loss": 0.0017,
      "epoch": 2.4052980132450332,
      "grad_norm": 0.5302432179450989,
      "learning_rate": 2.9856701445398016e-05,
      "lm_loss": 0.0007,
      "loss": 0.0025,
      "step": 454
    },
    {
      "classification_loss": 0.0012,
      "epoch": 2.410596026490066,
      "grad_norm": 0.05441528186202049,
      "learning_rate": 2.9802790845215517e-05,
      "lm_loss": 0.0002,
      "loss": 0.0013,
      "step": 455
    },
    {
      "classification_loss": 0.0004,
      "epoch": 2.4158940397350994,
      "grad_norm": 0.01915786601603031,
      "learning_rate": 2.974878633390639e-05,
      "lm_loss": 0.0001,
      "loss": 0.0005,
      "step": 456
    },
    {
      "classification_loss": 0.0009,
      "epoch": 2.4211920529801323,
      "grad_norm": 0.07995589077472687,
      "learning_rate": 2.9694688428836006e-05,
      "lm_loss": 0.0002,
      "loss": 0.0011,
      "step": 457
    },
    {
      "classification_loss": 0.2881,
      "epoch": 2.4264900662251656,
      "grad_norm": 16.447843551635742,
      "learning_rate": 2.9640497648264442e-05,
      "lm_loss": 0.0237,
      "loss": 0.3119,
      "step": 458
    },
    {
      "classification_loss": 0.0004,
      "epoch": 2.4317880794701985,
      "grad_norm": 0.01886656880378723,
      "learning_rate": 2.9586214511341542e-05,
      "lm_loss": 0.0001,
      "loss": 0.0005,
      "step": 459
    },
    {
      "classification_loss": 0.0008,
      "epoch": 2.437086092715232,
      "grad_norm": 0.04251056909561157,
      "learning_rate": 2.953183953810192e-05,
      "lm_loss": 0.0002,
      "loss": 0.001,
      "step": 460
    },
    {
      "classification_loss": 0.0004,
      "epoch": 2.442384105960265,
      "grad_norm": 0.022392399609088898,
      "learning_rate": 2.9477373249459974e-05,
      "lm_loss": 0.0001,
      "loss": 0.0005,
      "step": 461
    },
    {
      "classification_loss": 0.0007,
      "epoch": 2.447682119205298,
      "grad_norm": 0.03564208000898361,
      "learning_rate": 2.942281616720493e-05,
      "lm_loss": 0.0002,
      "loss": 0.0009,
      "step": 462
    },
    {
      "classification_loss": 0.0006,
      "epoch": 2.4529801324503313,
      "grad_norm": 0.3286847472190857,
      "learning_rate": 2.9368168813995806e-05,
      "lm_loss": 0.0007,
      "loss": 0.0013,
      "step": 463
    },
    {
      "classification_loss": 0.0005,
      "epoch": 2.458278145695364,
      "grad_norm": 0.028327343985438347,
      "learning_rate": 2.9313431713356414e-05,
      "lm_loss": 0.0001,
      "loss": 0.0006,
      "step": 464
    },
    {
      "classification_loss": 0.0004,
      "epoch": 2.4635761589403975,
      "grad_norm": 0.04668082669377327,
      "learning_rate": 2.9258605389670367e-05,
      "lm_loss": 0.0001,
      "loss": 0.0005,
      "step": 465
    },
    {
      "classification_loss": 0.3739,
      "epoch": 2.4688741721854304,
      "grad_norm": 41.33816909790039,
      "learning_rate": 2.9203690368176026e-05,
      "lm_loss": 0.0269,
      "loss": 0.4007,
      "step": 466
    },
    {
      "classification_loss": 0.0006,
      "epoch": 2.4741721854304637,
      "grad_norm": 0.11546545475721359,
      "learning_rate": 2.9148687174961478e-05,
      "lm_loss": 0.0002,
      "loss": 0.0008,
      "step": 467
    },
    {
      "classification_loss": 0.0003,
      "epoch": 2.4794701986754966,
      "grad_norm": 0.014606965705752373,
      "learning_rate": 2.9093596336959513e-05,
      "lm_loss": 0.0001,
      "loss": 0.0004,
      "step": 468
    },
    {
      "classification_loss": 0.0387,
      "epoch": 2.48476821192053,
      "grad_norm": 26.499374389648438,
      "learning_rate": 2.9038418381942547e-05,
      "lm_loss": 0.0049,
      "loss": 0.0437,
      "step": 469
    },
    {
      "classification_loss": 0.001,
      "epoch": 2.4900662251655628,
      "grad_norm": 0.0944555476307869,
      "learning_rate": 2.8983153838517588e-05,
      "lm_loss": 0.0002,
      "loss": 0.0011,
      "step": 470
    },
    {
      "classification_loss": 0.2056,
      "epoch": 2.495364238410596,
      "grad_norm": 72.76007080078125,
      "learning_rate": 2.8927803236121153e-05,
      "lm_loss": 0.0171,
      "loss": 0.2226,
      "step": 471
    },
    {
      "classification_loss": 0.0003,
      "epoch": 2.500662251655629,
      "grad_norm": 0.013602725230157375,
      "learning_rate": 2.8872367105014235e-05,
      "lm_loss": 0.0,
      "loss": 0.0003,
      "step": 472
    },
    {
      "classification_loss": 0.0002,
      "epoch": 2.5059602649006623,
      "grad_norm": 0.008087028749287128,
      "learning_rate": 2.881684597627717e-05,
      "lm_loss": 0.0,
      "loss": 0.0002,
      "step": 473
    },
    {
      "classification_loss": 0.0002,
      "epoch": 2.5112582781456956,
      "grad_norm": 0.020297549664974213,
      "learning_rate": 2.876124038180457e-05,
      "lm_loss": 0.0001,
      "loss": 0.0003,
      "step": 474
    },
    {
      "classification_loss": 0.0001,
      "epoch": 2.5165562913907285,
      "grad_norm": 0.00973586831241846,
      "learning_rate": 2.8705550854300268e-05,
      "lm_loss": 0.0001,
      "loss": 0.0002,
      "step": 475
    },
    {
      "classification_loss": 0.0001,
      "epoch": 2.5218543046357613,
      "grad_norm": 0.007681860122829676,
      "learning_rate": 2.8649777927272136e-05,
      "lm_loss": 0.0,
      "loss": 0.0002,
      "step": 476
    },
    {
      "classification_loss": 0.001,
      "epoch": 2.5271523178807946,
      "grad_norm": 2.2136242389678955,
      "learning_rate": 2.8593922135027038e-05,
      "lm_loss": 0.0006,
      "loss": 0.0016,
      "step": 477
    },
    {
      "classification_loss": 0.0006,
      "epoch": 2.532450331125828,
      "grad_norm": 0.028420673683285713,
      "learning_rate": 2.8537984012665694e-05,
      "lm_loss": 0.0001,
      "loss": 0.0007,
      "step": 478
    },
    {
      "classification_loss": 0.0001,
      "epoch": 2.537748344370861,
      "grad_norm": 0.006430044770240784,
      "learning_rate": 2.848196409607754e-05,
      "lm_loss": 0.0,
      "loss": 0.0002,
      "step": 479
    },
    {
      "classification_loss": 0.0002,
      "epoch": 2.543046357615894,
      "grad_norm": 0.03874123468995094,
      "learning_rate": 2.8425862921935604e-05,
      "lm_loss": 0.0001,
      "loss": 0.0002,
      "step": 480
    },
    {
      "classification_loss": 0.0011,
      "epoch": 2.548344370860927,
      "grad_norm": 0.5411393642425537,
      "learning_rate": 2.836968102769138e-05,
      "lm_loss": 0.0006,
      "loss": 0.0017,
      "step": 481
    },
    {
      "classification_loss": 0.0022,
      "epoch": 2.5536423841059603,
      "grad_norm": 0.5700423717498779,
      "learning_rate": 2.831341895156964e-05,
      "lm_loss": 0.0008,
      "loss": 0.003,
      "step": 482
    },
    {
      "classification_loss": 0.0009,
      "epoch": 2.558940397350993,
      "grad_norm": 0.3711094558238983,
      "learning_rate": 2.825707723256332e-05,
      "lm_loss": 0.0003,
      "loss": 0.0012,
      "step": 483
    },
    {
      "classification_loss": 0.0002,
      "epoch": 2.5642384105960265,
      "grad_norm": 0.0275566503405571,
      "learning_rate": 2.8200656410428323e-05,
      "lm_loss": 0.0,
      "loss": 0.0003,
      "step": 484
    },
    {
      "classification_loss": 0.5089,
      "epoch": 2.5695364238410594,
      "grad_norm": 12.725303649902344,
      "learning_rate": 2.8144157025678383e-05,
      "lm_loss": 0.0417,
      "loss": 0.5506,
      "step": 485
    },
    {
      "classification_loss": 0.0001,
      "epoch": 2.5748344370860927,
      "grad_norm": 0.008202818222343922,
      "learning_rate": 2.808757961957985e-05,
      "lm_loss": 0.0,
      "loss": 0.0002,
      "step": 486
    },
    {
      "classification_loss": 0.0002,
      "epoch": 2.580132450331126,
      "grad_norm": 0.041455235332250595,
      "learning_rate": 2.8030924734146524e-05,
      "lm_loss": 0.0001,
      "loss": 0.0003,
      "step": 487
    },
    {
      "classification_loss": 0.0002,
      "epoch": 2.585430463576159,
      "grad_norm": 0.008234069682657719,
      "learning_rate": 2.7974192912134472e-05,
      "lm_loss": 0.0,
      "loss": 0.0002,
      "step": 488
    },
    {
      "classification_loss": 0.0001,
      "epoch": 2.590728476821192,
      "grad_norm": 0.00477932533249259,
      "learning_rate": 2.7917384697036803e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 489
    },
    {
      "classification_loss": 0.0004,
      "epoch": 2.596026490066225,
      "grad_norm": 0.17319636046886444,
      "learning_rate": 2.7860500633078475e-05,
      "lm_loss": 0.0002,
      "loss": 0.0005,
      "step": 490
    },
    {
      "classification_loss": 0.0001,
      "epoch": 2.6013245033112584,
      "grad_norm": 0.006443544756621122,
      "learning_rate": 2.7803541265211092e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 491
    },
    {
      "classification_loss": 0.0007,
      "epoch": 2.6066225165562913,
      "grad_norm": 0.44631409645080566,
      "learning_rate": 2.7746507139107667e-05,
      "lm_loss": 0.0002,
      "loss": 0.0009,
      "step": 492
    },
    {
      "classification_loss": 0.0001,
      "epoch": 2.6119205298013246,
      "grad_norm": 0.004931404255330563,
      "learning_rate": 2.7689398801157393e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 493
    },
    {
      "classification_loss": 0.0002,
      "epoch": 2.6172185430463575,
      "grad_norm": 0.009196000173687935,
      "learning_rate": 2.7632216798460426e-05,
      "lm_loss": 0.0,
      "loss": 0.0002,
      "step": 494
    },
    {
      "classification_loss": 0.0002,
      "epoch": 2.622516556291391,
      "grad_norm": 0.010494556277990341,
      "learning_rate": 2.757496167882263e-05,
      "lm_loss": 0.0,
      "loss": 0.0002,
      "step": 495
    },
    {
      "classification_loss": 0.0001,
      "epoch": 2.627814569536424,
      "grad_norm": 0.0073355198837816715,
      "learning_rate": 2.7517633990750328e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 496
    },
    {
      "classification_loss": 0.0002,
      "epoch": 2.633112582781457,
      "grad_norm": 0.0110849654302001,
      "learning_rate": 2.7460234283445052e-05,
      "lm_loss": 0.0,
      "loss": 0.0002,
      "step": 497
    },
    {
      "classification_loss": 0.0001,
      "epoch": 2.63841059602649,
      "grad_norm": 0.007363836746662855,
      "learning_rate": 2.7402763106798295e-05,
      "lm_loss": 0.0,
      "loss": 0.0002,
      "step": 498
    },
    {
      "classification_loss": 0.0002,
      "epoch": 2.643708609271523,
      "grad_norm": 0.008405833505094051,
      "learning_rate": 2.7345221011386197e-05,
      "lm_loss": 0.0,
      "loss": 0.0002,
      "step": 499
    },
    {
      "classification_loss": 0.0002,
      "epoch": 2.6490066225165565,
      "grad_norm": 0.014074385166168213,
      "learning_rate": 2.728760854846434e-05,
      "lm_loss": 0.0,
      "loss": 0.0003,
      "step": 500
    },
    {
      "epoch": 2.6490066225165565,
      "eval_accuracy": 0.9126760363578796,
      "eval_auroc": 0.9614749550819397,
      "eval_classification_loss": 0.6585378646850586,
      "eval_f1": 0.9051987528800964,
      "eval_lm_loss": 0.06123493239283562,
      "eval_loss": 0.7197728157043457,
      "eval_precision": 0.9610389471054077,
      "eval_recall": 0.8554913401603699,
      "eval_runtime": 59.0852,
      "eval_samples_per_second": 6.008,
      "eval_steps_per_second": 3.013,
      "step": 500
    },
    {
      "classification_loss": 0.0001,
      "epoch": 2.6543046357615894,
      "grad_norm": 0.007277980446815491,
      "learning_rate": 2.7229926269962414e-05,
      "lm_loss": 0.0,
      "loss": 0.0002,
      "step": 501
    },
    {
      "classification_loss": 0.0001,
      "epoch": 2.6596026490066222,
      "grad_norm": 0.007178813219070435,
      "learning_rate": 2.7172174728478932e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 502
    },
    {
      "classification_loss": 0.9097,
      "epoch": 2.6649006622516556,
      "grad_norm": 18.918231964111328,
      "learning_rate": 2.7114354477275974e-05,
      "lm_loss": 0.0907,
      "loss": 1.0003,
      "step": 503
    },
    {
      "classification_loss": 0.0001,
      "epoch": 2.670198675496689,
      "grad_norm": 0.005723870825022459,
      "learning_rate": 2.705646607027384e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 504
    },
    {
      "classification_loss": 0.0002,
      "epoch": 2.6754966887417218,
      "grad_norm": 0.011710802093148232,
      "learning_rate": 2.6998510062045773e-05,
      "lm_loss": 0.0,
      "loss": 0.0003,
      "step": 505
    },
    {
      "classification_loss": 0.0002,
      "epoch": 2.680794701986755,
      "grad_norm": 0.009104592725634575,
      "learning_rate": 2.694048700781264e-05,
      "lm_loss": 0.0,
      "loss": 0.0002,
      "step": 506
    },
    {
      "classification_loss": 0.5998,
      "epoch": 2.686092715231788,
      "grad_norm": 7.0369367599487305,
      "learning_rate": 2.6882397463437615e-05,
      "lm_loss": 0.0626,
      "loss": 0.6624,
      "step": 507
    },
    {
      "classification_loss": 0.0002,
      "epoch": 2.6913907284768213,
      "grad_norm": 0.014677820727229118,
      "learning_rate": 2.682424198542085e-05,
      "lm_loss": 0.0001,
      "loss": 0.0002,
      "step": 508
    },
    {
      "classification_loss": 0.0004,
      "epoch": 2.6966887417218546,
      "grad_norm": 0.02184452675282955,
      "learning_rate": 2.6766021130894134e-05,
      "lm_loss": 0.0001,
      "loss": 0.0004,
      "step": 509
    },
    {
      "classification_loss": 0.0002,
      "epoch": 2.7019867549668874,
      "grad_norm": 0.014110983349382877,
      "learning_rate": 2.6707735457615575e-05,
      "lm_loss": 0.0,
      "loss": 0.0002,
      "step": 510
    },
    {
      "classification_loss": 0.3236,
      "epoch": 2.7072847682119203,
      "grad_norm": 21.53240966796875,
      "learning_rate": 2.6649385523964232e-05,
      "lm_loss": 0.0233,
      "loss": 0.3469,
      "step": 511
    },
    {
      "classification_loss": 0.0003,
      "epoch": 2.7125827814569536,
      "grad_norm": 0.01429409347474575,
      "learning_rate": 2.659097188893481e-05,
      "lm_loss": 0.0001,
      "loss": 0.0003,
      "step": 512
    },
    {
      "classification_loss": 0.0006,
      "epoch": 2.717880794701987,
      "grad_norm": 0.059666700661182404,
      "learning_rate": 2.6532495112132244e-05,
      "lm_loss": 0.0001,
      "loss": 0.0007,
      "step": 513
    },
    {
      "classification_loss": 0.0337,
      "epoch": 2.72317880794702,
      "grad_norm": 10.775862693786621,
      "learning_rate": 2.6473955753766398e-05,
      "lm_loss": 0.0024,
      "loss": 0.0361,
      "step": 514
    },
    {
      "classification_loss": 0.0003,
      "epoch": 2.7284768211920527,
      "grad_norm": 0.01527821272611618,
      "learning_rate": 2.641535437464665e-05,
      "lm_loss": 0.0001,
      "loss": 0.0004,
      "step": 515
    },
    {
      "classification_loss": 0.0006,
      "epoch": 2.733774834437086,
      "grad_norm": 0.07742226123809814,
      "learning_rate": 2.6356691536176538e-05,
      "lm_loss": 0.0001,
      "loss": 0.0007,
      "step": 516
    },
    {
      "classification_loss": 0.2339,
      "epoch": 2.7390728476821193,
      "grad_norm": 31.22527503967285,
      "learning_rate": 2.62979678003484e-05,
      "lm_loss": 0.0195,
      "loss": 0.2534,
      "step": 517
    },
    {
      "classification_loss": 0.0005,
      "epoch": 2.744370860927152,
      "grad_norm": 0.02605348452925682,
      "learning_rate": 2.6239183729737957e-05,
      "lm_loss": 0.0001,
      "loss": 0.0006,
      "step": 518
    },
    {
      "classification_loss": 0.0035,
      "epoch": 2.7496688741721855,
      "grad_norm": 2.1541953086853027,
      "learning_rate": 2.618033988749895e-05,
      "lm_loss": 0.0011,
      "loss": 0.0046,
      "step": 519
    },
    {
      "classification_loss": 0.0006,
      "epoch": 2.7549668874172184,
      "grad_norm": 0.035049907863140106,
      "learning_rate": 2.6121436837357733e-05,
      "lm_loss": 0.0001,
      "loss": 0.0007,
      "step": 520
    },
    {
      "classification_loss": 0.0012,
      "epoch": 2.7602649006622517,
      "grad_norm": 0.6209406852722168,
      "learning_rate": 2.6062475143607868e-05,
      "lm_loss": 0.0003,
      "loss": 0.0015,
      "step": 521
    },
    {
      "classification_loss": 0.0005,
      "epoch": 2.765562913907285,
      "grad_norm": 0.022160859778523445,
      "learning_rate": 2.6003455371104727e-05,
      "lm_loss": 0.0001,
      "loss": 0.0005,
      "step": 522
    },
    {
      "classification_loss": 0.0004,
      "epoch": 2.770860927152318,
      "grad_norm": 0.01831023022532463,
      "learning_rate": 2.594437808526009e-05,
      "lm_loss": 0.0001,
      "loss": 0.0004,
      "step": 523
    },
    {
      "classification_loss": 0.3139,
      "epoch": 2.776158940397351,
      "grad_norm": 23.14295196533203,
      "learning_rate": 2.588524385203669e-05,
      "lm_loss": 0.0277,
      "loss": 0.3415,
      "step": 524
    },
    {
      "classification_loss": 0.0005,
      "epoch": 2.781456953642384,
      "grad_norm": 0.0621364563703537,
      "learning_rate": 2.582605323794286e-05,
      "lm_loss": 0.0002,
      "loss": 0.0007,
      "step": 525
    },
    {
      "classification_loss": 0.0013,
      "epoch": 2.7867549668874174,
      "grad_norm": 0.9596188068389893,
      "learning_rate": 2.576680681002702e-05,
      "lm_loss": 0.0004,
      "loss": 0.0017,
      "step": 526
    },
    {
      "classification_loss": 0.0008,
      "epoch": 2.7920529801324503,
      "grad_norm": 0.06905034929513931,
      "learning_rate": 2.570750513587231e-05,
      "lm_loss": 0.0002,
      "loss": 0.0009,
      "step": 527
    },
    {
      "classification_loss": 0.0008,
      "epoch": 2.7973509933774836,
      "grad_norm": 0.12334898114204407,
      "learning_rate": 2.5648148783591142e-05,
      "lm_loss": 0.0002,
      "loss": 0.001,
      "step": 528
    },
    {
      "classification_loss": 0.0011,
      "epoch": 2.8026490066225165,
      "grad_norm": 0.052833814173936844,
      "learning_rate": 2.5588738321819724e-05,
      "lm_loss": 0.0001,
      "loss": 0.0012,
      "step": 529
    },
    {
      "classification_loss": 0.0006,
      "epoch": 2.80794701986755,
      "grad_norm": 0.029968678951263428,
      "learning_rate": 2.5529274319712656e-05,
      "lm_loss": 0.0001,
      "loss": 0.0007,
      "step": 530
    },
    {
      "classification_loss": 0.0005,
      "epoch": 2.8132450331125827,
      "grad_norm": 0.07342352718114853,
      "learning_rate": 2.546975734693744e-05,
      "lm_loss": 0.0001,
      "loss": 0.0006,
      "step": 531
    },
    {
      "classification_loss": 0.0005,
      "epoch": 2.818543046357616,
      "grad_norm": 0.030646558851003647,
      "learning_rate": 2.5410187973669035e-05,
      "lm_loss": 0.0001,
      "loss": 0.0006,
      "step": 532
    },
    {
      "classification_loss": 0.0003,
      "epoch": 2.823841059602649,
      "grad_norm": 0.012134003452956676,
      "learning_rate": 2.5350566770584423e-05,
      "lm_loss": 0.0,
      "loss": 0.0003,
      "step": 533
    },
    {
      "classification_loss": 0.2366,
      "epoch": 2.829139072847682,
      "grad_norm": 55.83869552612305,
      "learning_rate": 2.529089430885709e-05,
      "lm_loss": 0.0209,
      "loss": 0.2575,
      "step": 534
    },
    {
      "classification_loss": 0.0003,
      "epoch": 2.8344370860927155,
      "grad_norm": 0.01527783926576376,
      "learning_rate": 2.523117116015161e-05,
      "lm_loss": 0.0,
      "loss": 0.0003,
      "step": 535
    },
    {
      "classification_loss": 0.6699,
      "epoch": 2.8397350993377484,
      "grad_norm": 66.58013153076172,
      "learning_rate": 2.517139789661812e-05,
      "lm_loss": 0.0439,
      "loss": 0.7138,
      "step": 536
    },
    {
      "classification_loss": 0.0004,
      "epoch": 2.8450331125827812,
      "grad_norm": 0.020805561915040016,
      "learning_rate": 2.511157509088687e-05,
      "lm_loss": 0.0001,
      "loss": 0.0005,
      "step": 537
    },
    {
      "classification_loss": 0.747,
      "epoch": 2.8503311258278146,
      "grad_norm": 29.242202758789062,
      "learning_rate": 2.50517033160627e-05,
      "lm_loss": 0.0633,
      "loss": 0.8102,
      "step": 538
    },
    {
      "classification_loss": 0.0004,
      "epoch": 2.855629139072848,
      "grad_norm": 0.019773608073592186,
      "learning_rate": 2.4991783145719625e-05,
      "lm_loss": 0.0,
      "loss": 0.0005,
      "step": 539
    },
    {
      "classification_loss": 0.0003,
      "epoch": 2.8609271523178808,
      "grad_norm": 0.014965825714170933,
      "learning_rate": 2.4931815153895246e-05,
      "lm_loss": 0.0,
      "loss": 0.0003,
      "step": 540
    },
    {
      "classification_loss": 0.0004,
      "epoch": 2.866225165562914,
      "grad_norm": 0.02509867586195469,
      "learning_rate": 2.4871799915085314e-05,
      "lm_loss": 0.0001,
      "loss": 0.0005,
      "step": 541
    },
    {
      "classification_loss": 0.0003,
      "epoch": 2.871523178807947,
      "grad_norm": 0.033704496920108795,
      "learning_rate": 2.481173800423821e-05,
      "lm_loss": 0.0001,
      "loss": 0.0004,
      "step": 542
    },
    {
      "classification_loss": 0.0005,
      "epoch": 2.8768211920529803,
      "grad_norm": 0.028245626017451286,
      "learning_rate": 2.4751629996749427e-05,
      "lm_loss": 0.0001,
      "loss": 0.0005,
      "step": 543
    },
    {
      "classification_loss": 0.0006,
      "epoch": 2.882119205298013,
      "grad_norm": 0.10853353142738342,
      "learning_rate": 2.4691476468456074e-05,
      "lm_loss": 0.0001,
      "loss": 0.0008,
      "step": 544
    },
    {
      "classification_loss": 0.3745,
      "epoch": 2.8874172185430464,
      "grad_norm": 11.916918754577637,
      "learning_rate": 2.4631277995631352e-05,
      "lm_loss": 0.0314,
      "loss": 0.4059,
      "step": 545
    },
    {
      "classification_loss": 0.0008,
      "epoch": 2.8927152317880793,
      "grad_norm": 0.0394030325114727,
      "learning_rate": 2.4571035154979014e-05,
      "lm_loss": 0.0001,
      "loss": 0.0008,
      "step": 546
    },
    {
      "classification_loss": 0.3728,
      "epoch": 2.8980132450331126,
      "grad_norm": 6.1080403327941895,
      "learning_rate": 2.4510748523627887e-05,
      "lm_loss": 0.0461,
      "loss": 0.4189,
      "step": 547
    },
    {
      "classification_loss": 0.0005,
      "epoch": 2.903311258278146,
      "grad_norm": 0.023642556741833687,
      "learning_rate": 2.445041867912629e-05,
      "lm_loss": 0.0,
      "loss": 0.0005,
      "step": 548
    },
    {
      "classification_loss": 0.0004,
      "epoch": 2.908609271523179,
      "grad_norm": 0.022078802809119225,
      "learning_rate": 2.439004619943654e-05,
      "lm_loss": 0.0001,
      "loss": 0.0004,
      "step": 549
    },
    {
      "classification_loss": 0.0006,
      "epoch": 2.9139072847682117,
      "grad_norm": 0.03306747227907181,
      "learning_rate": 2.4329631662929397e-05,
      "lm_loss": 0.0001,
      "loss": 0.0007,
      "step": 550
    },
    {
      "epoch": 2.9139072847682117,
      "eval_accuracy": 0.9154929518699646,
      "eval_auroc": 0.9633487462997437,
      "eval_classification_loss": 0.446094810962677,
      "eval_f1": 0.9147727489471436,
      "eval_lm_loss": 0.04339073598384857,
      "eval_loss": 0.4894855320453644,
      "eval_precision": 0.8994413614273071,
      "eval_recall": 0.9306358098983765,
      "eval_runtime": 61.3893,
      "eval_samples_per_second": 5.783,
      "eval_steps_per_second": 2.9,
      "step": 550
    },
    {
      "classification_loss": 0.1345,
      "epoch": 2.919205298013245,
      "grad_norm": 50.191158294677734,
      "learning_rate": 2.4269175648378523e-05,
      "lm_loss": 0.0086,
      "loss": 0.143,
      "step": 551
    },
    {
      "classification_loss": 0.1213,
      "epoch": 2.9245033112582783,
      "grad_norm": 19.398170471191406,
      "learning_rate": 2.420867873495494e-05,
      "lm_loss": 0.0088,
      "loss": 0.13,
      "step": 552
    },
    {
      "classification_loss": 0.0099,
      "epoch": 2.929801324503311,
      "grad_norm": 2.329904317855835,
      "learning_rate": 2.4148141502221503e-05,
      "lm_loss": 0.0017,
      "loss": 0.0116,
      "step": 553
    },
    {
      "classification_loss": 0.0086,
      "epoch": 2.9350993377483445,
      "grad_norm": 6.2573323249816895,
      "learning_rate": 2.40875645301273e-05,
      "lm_loss": 0.0024,
      "loss": 0.011,
      "step": 554
    },
    {
      "classification_loss": 0.0011,
      "epoch": 2.9403973509933774,
      "grad_norm": 0.06344883143901825,
      "learning_rate": 2.402694839900214e-05,
      "lm_loss": 0.0001,
      "loss": 0.0012,
      "step": 555
    },
    {
      "classification_loss": 0.0064,
      "epoch": 2.9456953642384107,
      "grad_norm": 1.8969309329986572,
      "learning_rate": 2.3966293689550968e-05,
      "lm_loss": 0.0016,
      "loss": 0.008,
      "step": 556
    },
    {
      "classification_loss": 0.0015,
      "epoch": 2.9509933774834436,
      "grad_norm": 0.13037782907485962,
      "learning_rate": 2.3905600982848318e-05,
      "lm_loss": 0.0003,
      "loss": 0.0017,
      "step": 557
    },
    {
      "classification_loss": 0.0034,
      "epoch": 2.956291390728477,
      "grad_norm": 1.2463749647140503,
      "learning_rate": 2.384487086033274e-05,
      "lm_loss": 0.0014,
      "loss": 0.0048,
      "step": 558
    },
    {
      "classification_loss": 0.0009,
      "epoch": 2.96158940397351,
      "grad_norm": 0.24586497247219086,
      "learning_rate": 2.3784103903801227e-05,
      "lm_loss": 0.0002,
      "loss": 0.0011,
      "step": 559
    },
    {
      "classification_loss": 0.0594,
      "epoch": 2.966887417218543,
      "grad_norm": 36.230960845947266,
      "learning_rate": 2.372330069540364e-05,
      "lm_loss": 0.0066,
      "loss": 0.0659,
      "step": 560
    },
    {
      "classification_loss": 0.0201,
      "epoch": 2.9721854304635764,
      "grad_norm": 6.873016834259033,
      "learning_rate": 2.3662461817637125e-05,
      "lm_loss": 0.002,
      "loss": 0.0221,
      "step": 561
    },
    {
      "classification_loss": 0.0004,
      "epoch": 2.9774834437086093,
      "grad_norm": 0.02800923027098179,
      "learning_rate": 2.360158785334057e-05,
      "lm_loss": 0.0,
      "loss": 0.0005,
      "step": 562
    },
    {
      "classification_loss": 0.0009,
      "epoch": 2.982781456953642,
      "grad_norm": 0.05778644233942032,
      "learning_rate": 2.3540679385688967e-05,
      "lm_loss": 0.0001,
      "loss": 0.001,
      "step": 563
    },
    {
      "classification_loss": 0.2477,
      "epoch": 2.9880794701986755,
      "grad_norm": 22.107051849365234,
      "learning_rate": 2.347973699818787e-05,
      "lm_loss": 0.0201,
      "loss": 0.2678,
      "step": 564
    },
    {
      "classification_loss": 0.4233,
      "epoch": 2.993377483443709,
      "grad_norm": 9.97690200805664,
      "learning_rate": 2.3418761274667777e-05,
      "lm_loss": 0.0372,
      "loss": 0.4605,
      "step": 565
    },
    {
      "classification_loss": 0.0006,
      "epoch": 2.9986754966887417,
      "grad_norm": 0.039230432361364365,
      "learning_rate": 2.3357752799278538e-05,
      "lm_loss": 0.0001,
      "loss": 0.0007,
      "step": 566
    },
    {
      "classification_loss": 0.4982,
      "epoch": 3.003973509933775,
      "grad_norm": 12.18735122680664,
      "learning_rate": 2.3296712156483783e-05,
      "lm_loss": 0.0495,
      "loss": 0.5477,
      "step": 567
    },
    {
      "classification_loss": 0.0011,
      "epoch": 3.009271523178808,
      "grad_norm": 0.09520389139652252,
      "learning_rate": 2.32356399310553e-05,
      "lm_loss": 0.0001,
      "loss": 0.0012,
      "step": 568
    },
    {
      "classification_loss": 0.0015,
      "epoch": 3.014569536423841,
      "grad_norm": 0.09453952312469482,
      "learning_rate": 2.3174536708067433e-05,
      "lm_loss": 0.0001,
      "loss": 0.0016,
      "step": 569
    },
    {
      "classification_loss": 0.0006,
      "epoch": 3.019867549668874,
      "grad_norm": 0.033111341297626495,
      "learning_rate": 2.31134030728915e-05,
      "lm_loss": 0.0,
      "loss": 0.0006,
      "step": 570
    },
    {
      "classification_loss": 0.0008,
      "epoch": 3.0251655629139074,
      "grad_norm": 0.043545644730329514,
      "learning_rate": 2.3052239611190168e-05,
      "lm_loss": 0.0001,
      "loss": 0.0008,
      "step": 571
    },
    {
      "classification_loss": 0.0012,
      "epoch": 3.0304635761589402,
      "grad_norm": 0.07852927595376968,
      "learning_rate": 2.299104690891182e-05,
      "lm_loss": 0.0001,
      "loss": 0.0012,
      "step": 572
    },
    {
      "classification_loss": 0.0014,
      "epoch": 3.0357615894039736,
      "grad_norm": 0.12363312393426895,
      "learning_rate": 2.2929825552284987e-05,
      "lm_loss": 0.0001,
      "loss": 0.0015,
      "step": 573
    },
    {
      "classification_loss": 0.0009,
      "epoch": 3.0410596026490064,
      "grad_norm": 0.05457129701972008,
      "learning_rate": 2.2868576127812702e-05,
      "lm_loss": 0.0001,
      "loss": 0.0009,
      "step": 574
    },
    {
      "classification_loss": 0.0009,
      "epoch": 3.0463576158940397,
      "grad_norm": 0.042348045855760574,
      "learning_rate": 2.2807299222266888e-05,
      "lm_loss": 0.0001,
      "loss": 0.0009,
      "step": 575
    },
    {
      "classification_loss": 0.0007,
      "epoch": 3.0516556291390726,
      "grad_norm": 0.04518580809235573,
      "learning_rate": 2.2745995422682748e-05,
      "lm_loss": 0.0001,
      "loss": 0.0008,
      "step": 576
    },
    {
      "classification_loss": 0.0005,
      "epoch": 3.056953642384106,
      "grad_norm": 0.02369764819741249,
      "learning_rate": 2.2684665316353112e-05,
      "lm_loss": 0.0001,
      "loss": 0.0005,
      "step": 577
    },
    {
      "classification_loss": 0.0012,
      "epoch": 3.0622516556291393,
      "grad_norm": 0.35181543231010437,
      "learning_rate": 2.262330949082285e-05,
      "lm_loss": 0.0002,
      "loss": 0.0014,
      "step": 578
    },
    {
      "classification_loss": 0.001,
      "epoch": 3.067549668874172,
      "grad_norm": 0.05868568271398544,
      "learning_rate": 2.2561928533883205e-05,
      "lm_loss": 0.0001,
      "loss": 0.0011,
      "step": 579
    },
    {
      "classification_loss": 0.0021,
      "epoch": 3.0728476821192054,
      "grad_norm": 0.5377780199050903,
      "learning_rate": 2.2500523033566203e-05,
      "lm_loss": 0.0005,
      "loss": 0.0026,
      "step": 580
    },
    {
      "classification_loss": 0.4183,
      "epoch": 3.0781456953642383,
      "grad_norm": 7.395751476287842,
      "learning_rate": 2.2439093578138964e-05,
      "lm_loss": 0.0425,
      "loss": 0.4608,
      "step": 581
    },
    {
      "classification_loss": 0.0008,
      "epoch": 3.0834437086092716,
      "grad_norm": 0.03938509523868561,
      "learning_rate": 2.2377640756098133e-05,
      "lm_loss": 0.0001,
      "loss": 0.0009,
      "step": 582
    },
    {
      "classification_loss": 0.0006,
      "epoch": 3.0887417218543045,
      "grad_norm": 0.03662256896495819,
      "learning_rate": 2.231616515616418e-05,
      "lm_loss": 0.0001,
      "loss": 0.0007,
      "step": 583
    },
    {
      "classification_loss": 0.0003,
      "epoch": 3.094039735099338,
      "grad_norm": 0.015934890136122704,
      "learning_rate": 2.225466736727581e-05,
      "lm_loss": 0.0,
      "loss": 0.0003,
      "step": 584
    },
    {
      "classification_loss": 0.0005,
      "epoch": 3.0993377483443707,
      "grad_norm": 0.03562748432159424,
      "learning_rate": 2.2193147978584277e-05,
      "lm_loss": 0.0001,
      "loss": 0.0006,
      "step": 585
    },
    {
      "classification_loss": 0.0006,
      "epoch": 3.104635761589404,
      "grad_norm": 0.04006592929363251,
      "learning_rate": 2.2131607579447784e-05,
      "lm_loss": 0.0001,
      "loss": 0.0007,
      "step": 586
    },
    {
      "classification_loss": 0.001,
      "epoch": 3.109933774834437,
      "grad_norm": 0.11585499346256256,
      "learning_rate": 2.2070046759425813e-05,
      "lm_loss": 0.0001,
      "loss": 0.0012,
      "step": 587
    },
    {
      "classification_loss": 0.0008,
      "epoch": 3.11523178807947,
      "grad_norm": 0.0398457795381546,
      "learning_rate": 2.200846610827345e-05,
      "lm_loss": 0.0001,
      "loss": 0.0008,
      "step": 588
    },
    {
      "classification_loss": 0.0018,
      "epoch": 3.120529801324503,
      "grad_norm": 0.3968581259250641,
      "learning_rate": 2.1946866215935802e-05,
      "lm_loss": 0.0004,
      "loss": 0.0022,
      "step": 589
    },
    {
      "classification_loss": 0.0016,
      "epoch": 3.1258278145695364,
      "grad_norm": 0.14592856168746948,
      "learning_rate": 2.1885247672542277e-05,
      "lm_loss": 0.0006,
      "loss": 0.0022,
      "step": 590
    },
    {
      "classification_loss": 0.001,
      "epoch": 3.1311258278145697,
      "grad_norm": 0.228903129696846,
      "learning_rate": 2.1823611068400986e-05,
      "lm_loss": 0.0003,
      "loss": 0.0013,
      "step": 591
    },
    {
      "classification_loss": 0.0005,
      "epoch": 3.1364238410596026,
      "grad_norm": 0.04328501597046852,
      "learning_rate": 2.1761956993993044e-05,
      "lm_loss": 0.0001,
      "loss": 0.0005,
      "step": 592
    },
    {
      "classification_loss": 0.0005,
      "epoch": 3.141721854304636,
      "grad_norm": 0.02685474045574665,
      "learning_rate": 2.170028603996695e-05,
      "lm_loss": 0.0001,
      "loss": 0.0006,
      "step": 593
    },
    {
      "classification_loss": 0.0005,
      "epoch": 3.147019867549669,
      "grad_norm": 0.025737935677170753,
      "learning_rate": 2.1638598797132882e-05,
      "lm_loss": 0.0001,
      "loss": 0.0005,
      "step": 594
    },
    {
      "classification_loss": 0.0002,
      "epoch": 3.152317880794702,
      "grad_norm": 0.012042353861033916,
      "learning_rate": 2.1576895856457093e-05,
      "lm_loss": 0.0,
      "loss": 0.0002,
      "step": 595
    },
    {
      "classification_loss": 0.0002,
      "epoch": 3.157615894039735,
      "grad_norm": 0.012236720882356167,
      "learning_rate": 2.1515177809056212e-05,
      "lm_loss": 0.0,
      "loss": 0.0003,
      "step": 596
    },
    {
      "classification_loss": 0.001,
      "epoch": 3.1629139072847683,
      "grad_norm": 0.23940910398960114,
      "learning_rate": 2.1453445246191583e-05,
      "lm_loss": 0.0002,
      "loss": 0.0012,
      "step": 597
    },
    {
      "classification_loss": 0.0002,
      "epoch": 3.168211920529801,
      "grad_norm": 0.014685581438243389,
      "learning_rate": 2.1391698759263625e-05,
      "lm_loss": 0.0,
      "loss": 0.0003,
      "step": 598
    },
    {
      "classification_loss": 0.0002,
      "epoch": 3.1735099337748345,
      "grad_norm": 0.016547564417123795,
      "learning_rate": 2.132993893980614e-05,
      "lm_loss": 0.0,
      "loss": 0.0003,
      "step": 599
    },
    {
      "classification_loss": 0.0003,
      "epoch": 3.1788079470198674,
      "grad_norm": 0.015439745970070362,
      "learning_rate": 2.1268166379480658e-05,
      "lm_loss": 0.0,
      "loss": 0.0003,
      "step": 600
    },
    {
      "epoch": 3.1788079470198674,
      "eval_accuracy": 0.8985915780067444,
      "eval_auroc": 0.9670964479446411,
      "eval_classification_loss": 0.5686928033828735,
      "eval_f1": 0.8934911489486694,
      "eval_lm_loss": 0.055827461183071136,
      "eval_loss": 0.6245202422142029,
      "eval_precision": 0.9151515364646912,
      "eval_recall": 0.8728323578834534,
      "eval_runtime": 65.7171,
      "eval_samples_per_second": 5.402,
      "eval_steps_per_second": 2.709,
      "step": 600
    },
    {
      "classification_loss": 0.0003,
      "epoch": 3.1841059602649007,
      "grad_norm": 0.020101506263017654,
      "learning_rate": 2.120638167007076e-05,
      "lm_loss": 0.0001,
      "loss": 0.0003,
      "step": 601
    },
    {
      "classification_loss": 0.0013,
      "epoch": 3.1894039735099335,
      "grad_norm": 0.2395239770412445,
      "learning_rate": 2.1144585403476426e-05,
      "lm_loss": 0.0003,
      "loss": 0.0016,
      "step": 602
    },
    {
      "classification_loss": 0.0005,
      "epoch": 3.194701986754967,
      "grad_norm": 0.03106399066746235,
      "learning_rate": 2.1082778171708355e-05,
      "lm_loss": 0.0001,
      "loss": 0.0005,
      "step": 603
    },
    {
      "classification_loss": 0.0002,
      "epoch": 3.2,
      "grad_norm": 0.01977233588695526,
      "learning_rate": 2.102096056688228e-05,
      "lm_loss": 0.0,
      "loss": 0.0003,
      "step": 604
    },
    {
      "classification_loss": 0.0002,
      "epoch": 3.205298013245033,
      "grad_norm": 0.013594997115433216,
      "learning_rate": 2.0959133181213325e-05,
      "lm_loss": 0.0,
      "loss": 0.0002,
      "step": 605
    },
    {
      "classification_loss": 0.0004,
      "epoch": 3.2105960264900664,
      "grad_norm": 0.05006929486989975,
      "learning_rate": 2.08972966070103e-05,
      "lm_loss": 0.0001,
      "loss": 0.0005,
      "step": 606
    },
    {
      "classification_loss": 0.0005,
      "epoch": 3.2158940397350992,
      "grad_norm": 0.08351855725049973,
      "learning_rate": 2.0835451436670055e-05,
      "lm_loss": 0.0001,
      "loss": 0.0005,
      "step": 607
    },
    {
      "classification_loss": 0.0001,
      "epoch": 3.2211920529801326,
      "grad_norm": 0.00827920250594616,
      "learning_rate": 2.077359826267178e-05,
      "lm_loss": 0.0,
      "loss": 0.0002,
      "step": 608
    },
    {
      "classification_loss": 0.0003,
      "epoch": 3.2264900662251654,
      "grad_norm": 0.02033345215022564,
      "learning_rate": 2.071173767757134e-05,
      "lm_loss": 0.0,
      "loss": 0.0003,
      "step": 609
    },
    {
      "classification_loss": 0.0004,
      "epoch": 3.2317880794701987,
      "grad_norm": 0.08960611373186111,
      "learning_rate": 2.0649870273995614e-05,
      "lm_loss": 0.0001,
      "loss": 0.0006,
      "step": 610
    },
    {
      "classification_loss": 0.0002,
      "epoch": 3.2370860927152316,
      "grad_norm": 0.014926898293197155,
      "learning_rate": 2.0587996644636782e-05,
      "lm_loss": 0.0,
      "loss": 0.0002,
      "step": 611
    },
    {
      "classification_loss": 0.0002,
      "epoch": 3.242384105960265,
      "grad_norm": 0.0171620212495327,
      "learning_rate": 2.0526117382246684e-05,
      "lm_loss": 0.0,
      "loss": 0.0003,
      "step": 612
    },
    {
      "classification_loss": 0.0002,
      "epoch": 3.247682119205298,
      "grad_norm": 0.015568455681204796,
      "learning_rate": 2.0464233079631116e-05,
      "lm_loss": 0.0,
      "loss": 0.0002,
      "step": 613
    },
    {
      "classification_loss": 0.0003,
      "epoch": 3.252980132450331,
      "grad_norm": 0.024862440302968025,
      "learning_rate": 2.040234432964416e-05,
      "lm_loss": 0.0,
      "loss": 0.0003,
      "step": 614
    },
    {
      "classification_loss": 0.0002,
      "epoch": 3.258278145695364,
      "grad_norm": 0.013776661828160286,
      "learning_rate": 2.0340451725182503e-05,
      "lm_loss": 0.0,
      "loss": 0.0003,
      "step": 615
    },
    {
      "classification_loss": 0.0002,
      "epoch": 3.2635761589403973,
      "grad_norm": 0.015745852142572403,
      "learning_rate": 2.027855585917977e-05,
      "lm_loss": 0.0,
      "loss": 0.0002,
      "step": 616
    },
    {
      "classification_loss": 0.0003,
      "epoch": 3.2688741721854306,
      "grad_norm": 0.1487286239862442,
      "learning_rate": 2.0216657324600798e-05,
      "lm_loss": 0.0001,
      "loss": 0.0004,
      "step": 617
    },
    {
      "classification_loss": 0.0001,
      "epoch": 3.2741721854304635,
      "grad_norm": 0.008310251869261265,
      "learning_rate": 2.0154756714436043e-05,
      "lm_loss": 0.0,
      "loss": 0.0002,
      "step": 618
    },
    {
      "classification_loss": 0.0001,
      "epoch": 3.279470198675497,
      "grad_norm": 0.007881787605583668,
      "learning_rate": 2.00928546216958e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 619
    },
    {
      "classification_loss": 0.0001,
      "epoch": 3.2847682119205297,
      "grad_norm": 0.009321430698037148,
      "learning_rate": 2.003095163940458e-05,
      "lm_loss": 0.0,
      "loss": 0.0002,
      "step": 620
    },
    {
      "classification_loss": 0.0001,
      "epoch": 3.290066225165563,
      "grad_norm": 0.01720263995230198,
      "learning_rate": 1.9969048360595423e-05,
      "lm_loss": 0.0,
      "loss": 0.0002,
      "step": 621
    },
    {
      "classification_loss": 0.0001,
      "epoch": 3.295364238410596,
      "grad_norm": 0.007219622377306223,
      "learning_rate": 1.990714537830421e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 622
    },
    {
      "classification_loss": 0.0005,
      "epoch": 3.300662251655629,
      "grad_norm": 0.17339986562728882,
      "learning_rate": 1.984524328556396e-05,
      "lm_loss": 0.0002,
      "loss": 0.0007,
      "step": 623
    },
    {
      "classification_loss": 0.0002,
      "epoch": 3.305960264900662,
      "grad_norm": 0.010375919751822948,
      "learning_rate": 1.9783342675399206e-05,
      "lm_loss": 0.0,
      "loss": 0.0002,
      "step": 624
    },
    {
      "classification_loss": 0.0002,
      "epoch": 3.3112582781456954,
      "grad_norm": 0.010349669493734837,
      "learning_rate": 1.972144414082024e-05,
      "lm_loss": 0.0,
      "loss": 0.0002,
      "step": 625
    },
    {
      "classification_loss": 0.0001,
      "epoch": 3.3165562913907287,
      "grad_norm": 0.007380208000540733,
      "learning_rate": 1.96595482748175e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 626
    },
    {
      "classification_loss": 0.0002,
      "epoch": 3.3218543046357616,
      "grad_norm": 0.009793221019208431,
      "learning_rate": 1.9597655670355848e-05,
      "lm_loss": 0.0,
      "loss": 0.0002,
      "step": 627
    },
    {
      "classification_loss": 0.0001,
      "epoch": 3.3271523178807945,
      "grad_norm": 0.008905098773539066,
      "learning_rate": 1.9535766920368888e-05,
      "lm_loss": 0.0,
      "loss": 0.0002,
      "step": 628
    },
    {
      "classification_loss": 0.0001,
      "epoch": 3.332450331125828,
      "grad_norm": 0.006328703835606575,
      "learning_rate": 1.947388261775332e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 629
    },
    {
      "classification_loss": 0.0001,
      "epoch": 3.337748344370861,
      "grad_norm": 0.006689694244414568,
      "learning_rate": 1.941200335536322e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 630
    },
    {
      "classification_loss": 0.0001,
      "epoch": 3.343046357615894,
      "grad_norm": 0.007170348893851042,
      "learning_rate": 1.9350129726004393e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 631
    },
    {
      "classification_loss": 0.0001,
      "epoch": 3.3483443708609273,
      "grad_norm": 0.0066578746773302555,
      "learning_rate": 1.9288262322428667e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 632
    },
    {
      "classification_loss": 0.0001,
      "epoch": 3.35364238410596,
      "grad_norm": 0.0038751934189349413,
      "learning_rate": 1.9226401737328224e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 633
    },
    {
      "classification_loss": 0.0001,
      "epoch": 3.3589403973509935,
      "grad_norm": 0.007950820028781891,
      "learning_rate": 1.916454856332995e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 634
    },
    {
      "classification_loss": 0.1149,
      "epoch": 3.3642384105960264,
      "grad_norm": 578.1722412109375,
      "learning_rate": 1.910270339298971e-05,
      "lm_loss": 0.014,
      "loss": 0.1289,
      "step": 635
    },
    {
      "classification_loss": 0.0002,
      "epoch": 3.3695364238410597,
      "grad_norm": 0.011052760295569897,
      "learning_rate": 1.904086681878668e-05,
      "lm_loss": 0.0,
      "loss": 0.0002,
      "step": 636
    },
    {
      "classification_loss": 0.0001,
      "epoch": 3.3748344370860925,
      "grad_norm": 0.005049990955740213,
      "learning_rate": 1.897903943311773e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 637
    },
    {
      "classification_loss": 0.0001,
      "epoch": 3.380132450331126,
      "grad_norm": 0.005164985544979572,
      "learning_rate": 1.8917221828291652e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 638
    },
    {
      "classification_loss": 0.0002,
      "epoch": 3.385430463576159,
      "grad_norm": 0.011179124005138874,
      "learning_rate": 1.8855414596523577e-05,
      "lm_loss": 0.0,
      "loss": 0.0002,
      "step": 639
    },
    {
      "classification_loss": 0.0002,
      "epoch": 3.390728476821192,
      "grad_norm": 0.015593500807881355,
      "learning_rate": 1.879361832992924e-05,
      "lm_loss": 0.0,
      "loss": 0.0002,
      "step": 640
    },
    {
      "classification_loss": 0.0001,
      "epoch": 3.396026490066225,
      "grad_norm": 0.008534195832908154,
      "learning_rate": 1.873183362051935e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 641
    },
    {
      "classification_loss": 0.0001,
      "epoch": 3.4013245033112582,
      "grad_norm": 0.00563429668545723,
      "learning_rate": 1.8670061060193867e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 642
    },
    {
      "classification_loss": 0.0001,
      "epoch": 3.4066225165562916,
      "grad_norm": 0.007203865330666304,
      "learning_rate": 1.8608301240736378e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 643
    },
    {
      "classification_loss": 0.0,
      "epoch": 3.4119205298013244,
      "grad_norm": 0.0023336398880928755,
      "learning_rate": 1.8546554753808424e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 644
    },
    {
      "classification_loss": 0.0,
      "epoch": 3.4172185430463577,
      "grad_norm": 0.0028586378321051598,
      "learning_rate": 1.8484822190943794e-05,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 645
    },
    {
      "classification_loss": 0.0001,
      "epoch": 3.4225165562913906,
      "grad_norm": 0.004268086049705744,
      "learning_rate": 1.842310414354291e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 646
    },
    {
      "classification_loss": 0.0002,
      "epoch": 3.427814569536424,
      "grad_norm": 0.05391693860292435,
      "learning_rate": 1.8361401202867128e-05,
      "lm_loss": 0.0001,
      "loss": 0.0003,
      "step": 647
    },
    {
      "classification_loss": 0.0001,
      "epoch": 3.433112582781457,
      "grad_norm": 0.0052588642574846745,
      "learning_rate": 1.8299713960033057e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 648
    },
    {
      "classification_loss": 0.0,
      "epoch": 3.43841059602649,
      "grad_norm": 0.002091035945340991,
      "learning_rate": 1.8238043006006963e-05,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 649
    },
    {
      "classification_loss": 0.0001,
      "epoch": 3.443708609271523,
      "grad_norm": 0.003742680186405778,
      "learning_rate": 1.8176388931599017e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 650
    },
    {
      "epoch": 3.443708609271523,
      "eval_accuracy": 0.9014084339141846,
      "eval_auroc": 0.9662072658538818,
      "eval_classification_loss": 0.687229573726654,
      "eval_f1": 0.8985507488250732,
      "eval_lm_loss": 0.06251314282417297,
      "eval_loss": 0.7497426867485046,
      "eval_precision": 0.9011628031730652,
      "eval_recall": 0.8959537744522095,
      "eval_runtime": 60.5664,
      "eval_samples_per_second": 5.861,
      "eval_steps_per_second": 2.939,
      "step": 650
    },
    {
      "classification_loss": 0.0001,
      "epoch": 3.4490066225165563,
      "grad_norm": 0.01925775408744812,
      "learning_rate": 1.8114752327457733e-05,
      "lm_loss": 0.0001,
      "loss": 0.0002,
      "step": 651
    },
    {
      "classification_loss": 0.0001,
      "epoch": 3.4543046357615896,
      "grad_norm": 0.005167665425688028,
      "learning_rate": 1.8053133784064208e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 652
    },
    {
      "classification_loss": 0.0001,
      "epoch": 3.4596026490066225,
      "grad_norm": 0.0035909651778638363,
      "learning_rate": 1.7991533891726555e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 653
    },
    {
      "classification_loss": 0.0001,
      "epoch": 3.4649006622516554,
      "grad_norm": 0.0036242862697690725,
      "learning_rate": 1.7929953240574197e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 654
    },
    {
      "classification_loss": 0.0,
      "epoch": 3.4701986754966887,
      "grad_norm": 0.002709699794650078,
      "learning_rate": 1.7868392420552215e-05,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 655
    },
    {
      "classification_loss": 0.0,
      "epoch": 3.475496688741722,
      "grad_norm": 0.0030914724338799715,
      "learning_rate": 1.780685202141573e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 656
    },
    {
      "classification_loss": 0.0,
      "epoch": 3.480794701986755,
      "grad_norm": 0.001309129991568625,
      "learning_rate": 1.7745332632724205e-05,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 657
    },
    {
      "classification_loss": 0.0,
      "epoch": 3.486092715231788,
      "grad_norm": 0.0031239136587828398,
      "learning_rate": 1.7683834843835826e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 658
    },
    {
      "classification_loss": 0.0,
      "epoch": 3.491390728476821,
      "grad_norm": 0.0019071531714871526,
      "learning_rate": 1.7622359243901874e-05,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 659
    },
    {
      "classification_loss": 0.0,
      "epoch": 3.4966887417218544,
      "grad_norm": 0.0020897502545267344,
      "learning_rate": 1.7560906421861035e-05,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 660
    },
    {
      "classification_loss": 0.0,
      "epoch": 3.5019867549668873,
      "grad_norm": 0.001477065379731357,
      "learning_rate": 1.7499476966433803e-05,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 661
    },
    {
      "classification_loss": 0.0003,
      "epoch": 3.5072847682119206,
      "grad_norm": 0.16811038553714752,
      "learning_rate": 1.7438071466116792e-05,
      "lm_loss": 0.0002,
      "loss": 0.0005,
      "step": 662
    },
    {
      "classification_loss": 0.5899,
      "epoch": 3.5125827814569535,
      "grad_norm": 7.528231143951416,
      "learning_rate": 1.7376690509177158e-05,
      "lm_loss": 0.0383,
      "loss": 0.6282,
      "step": 663
    },
    {
      "classification_loss": 0.0001,
      "epoch": 3.517880794701987,
      "grad_norm": 0.009575809352099895,
      "learning_rate": 1.7315334683646898e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 664
    },
    {
      "classification_loss": 0.0,
      "epoch": 3.52317880794702,
      "grad_norm": 0.0019360407022759318,
      "learning_rate": 1.725400457731726e-05,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 665
    },
    {
      "classification_loss": 0.0001,
      "epoch": 3.528476821192053,
      "grad_norm": 0.0058985017240047455,
      "learning_rate": 1.7192700777733115e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 666
    },
    {
      "classification_loss": 0.0006,
      "epoch": 3.533774834437086,
      "grad_norm": 0.1232844889163971,
      "learning_rate": 1.71314238721873e-05,
      "lm_loss": 0.0001,
      "loss": 0.0007,
      "step": 667
    },
    {
      "classification_loss": 0.0,
      "epoch": 3.539072847682119,
      "grad_norm": 0.0025346805341541767,
      "learning_rate": 1.707017444771502e-05,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 668
    },
    {
      "classification_loss": 0.0001,
      "epoch": 3.5443708609271525,
      "grad_norm": 0.0065282853320240974,
      "learning_rate": 1.7008953091088186e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 669
    },
    {
      "classification_loss": 0.1182,
      "epoch": 3.5496688741721854,
      "grad_norm": 179.83485412597656,
      "learning_rate": 1.6947760388809835e-05,
      "lm_loss": 0.0065,
      "loss": 0.1247,
      "step": 670
    },
    {
      "classification_loss": 0.0001,
      "epoch": 3.5549668874172187,
      "grad_norm": 0.0034845592454075813,
      "learning_rate": 1.6886596927108502e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 671
    },
    {
      "classification_loss": 0.1551,
      "epoch": 3.5602649006622515,
      "grad_norm": 13.291156768798828,
      "learning_rate": 1.6825463291932567e-05,
      "lm_loss": 0.0107,
      "loss": 0.1657,
      "step": 672
    },
    {
      "classification_loss": 0.0001,
      "epoch": 3.565562913907285,
      "grad_norm": 0.00786404125392437,
      "learning_rate": 1.6764360068944706e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 673
    },
    {
      "classification_loss": 0.0001,
      "epoch": 3.5708609271523177,
      "grad_norm": 0.008833058178424835,
      "learning_rate": 1.6703287843516227e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 674
    },
    {
      "classification_loss": 0.0001,
      "epoch": 3.576158940397351,
      "grad_norm": 0.011018514633178711,
      "learning_rate": 1.6642247200721465e-05,
      "lm_loss": 0.0,
      "loss": 0.0002,
      "step": 675
    },
    {
      "classification_loss": 0.0001,
      "epoch": 3.581456953642384,
      "grad_norm": 0.010098502039909363,
      "learning_rate": 1.6581238725332233e-05,
      "lm_loss": 0.0,
      "loss": 0.0002,
      "step": 676
    },
    {
      "classification_loss": 0.0001,
      "epoch": 3.5867549668874172,
      "grad_norm": 0.0036193369887769222,
      "learning_rate": 1.6520263001812133e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 677
    },
    {
      "classification_loss": 0.0001,
      "epoch": 3.5920529801324506,
      "grad_norm": 0.010048923082649708,
      "learning_rate": 1.6459320614311036e-05,
      "lm_loss": 0.0,
      "loss": 0.0002,
      "step": 678
    },
    {
      "classification_loss": 0.0001,
      "epoch": 3.5973509933774834,
      "grad_norm": 0.008332780562341213,
      "learning_rate": 1.6398412146659433e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 679
    },
    {
      "classification_loss": 0.0001,
      "epoch": 3.6026490066225163,
      "grad_norm": 0.00675076013430953,
      "learning_rate": 1.6337538182362878e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 680
    },
    {
      "classification_loss": 0.0002,
      "epoch": 3.6079470198675496,
      "grad_norm": 0.015578405000269413,
      "learning_rate": 1.6276699304596375e-05,
      "lm_loss": 0.0,
      "loss": 0.0002,
      "step": 681
    },
    {
      "classification_loss": 0.2766,
      "epoch": 3.613245033112583,
      "grad_norm": 23.013700485229492,
      "learning_rate": 1.6215896096198776e-05,
      "lm_loss": 0.0194,
      "loss": 0.296,
      "step": 682
    },
    {
      "classification_loss": 0.0005,
      "epoch": 3.618543046357616,
      "grad_norm": 0.08652503788471222,
      "learning_rate": 1.6155129139667266e-05,
      "lm_loss": 0.0001,
      "loss": 0.0006,
      "step": 683
    },
    {
      "classification_loss": 0.0003,
      "epoch": 3.623841059602649,
      "grad_norm": 0.027641918510198593,
      "learning_rate": 1.6094399017151682e-05,
      "lm_loss": 0.0,
      "loss": 0.0003,
      "step": 684
    },
    {
      "classification_loss": 0.0016,
      "epoch": 3.629139072847682,
      "grad_norm": 0.3050944209098816,
      "learning_rate": 1.603370631044904e-05,
      "lm_loss": 0.0001,
      "loss": 0.0017,
      "step": 685
    },
    {
      "classification_loss": 0.0003,
      "epoch": 3.6344370860927153,
      "grad_norm": 0.02661876380443573,
      "learning_rate": 1.597305160099787e-05,
      "lm_loss": 0.0,
      "loss": 0.0004,
      "step": 686
    },
    {
      "classification_loss": 0.0015,
      "epoch": 3.639735099337748,
      "grad_norm": 0.43420594930648804,
      "learning_rate": 1.5912435469872706e-05,
      "lm_loss": 0.0002,
      "loss": 0.0017,
      "step": 687
    },
    {
      "classification_loss": 0.001,
      "epoch": 3.6450331125827815,
      "grad_norm": 0.2760500907897949,
      "learning_rate": 1.58518584977785e-05,
      "lm_loss": 0.0004,
      "loss": 0.0013,
      "step": 688
    },
    {
      "classification_loss": 0.0005,
      "epoch": 3.6503311258278144,
      "grad_norm": 0.24832776188850403,
      "learning_rate": 1.5791321265045056e-05,
      "lm_loss": 0.0002,
      "loss": 0.0007,
      "step": 689
    },
    {
      "classification_loss": 0.0001,
      "epoch": 3.6556291390728477,
      "grad_norm": 0.0052548786625266075,
      "learning_rate": 1.5730824351621487e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 690
    },
    {
      "classification_loss": 0.0001,
      "epoch": 3.660927152317881,
      "grad_norm": 0.007747793570160866,
      "learning_rate": 1.5670368337070613e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 691
    },
    {
      "classification_loss": 0.0002,
      "epoch": 3.666225165562914,
      "grad_norm": 0.019217627122998238,
      "learning_rate": 1.5609953800563466e-05,
      "lm_loss": 0.0,
      "loss": 0.0003,
      "step": 692
    },
    {
      "classification_loss": 0.0002,
      "epoch": 3.6715231788079468,
      "grad_norm": 0.013566533103585243,
      "learning_rate": 1.5549581320873715e-05,
      "lm_loss": 0.0,
      "loss": 0.0002,
      "step": 693
    },
    {
      "classification_loss": 0.0006,
      "epoch": 3.67682119205298,
      "grad_norm": 0.19546669721603394,
      "learning_rate": 1.548925147637212e-05,
      "lm_loss": 0.0001,
      "loss": 0.0007,
      "step": 694
    },
    {
      "classification_loss": 0.2214,
      "epoch": 3.6821192052980134,
      "grad_norm": 20.25937843322754,
      "learning_rate": 1.542896484502099e-05,
      "lm_loss": 0.0212,
      "loss": 0.2426,
      "step": 695
    },
    {
      "classification_loss": 0.0001,
      "epoch": 3.6874172185430463,
      "grad_norm": 0.0047122105024755,
      "learning_rate": 1.536872200436865e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 696
    },
    {
      "classification_loss": 0.0001,
      "epoch": 3.6927152317880796,
      "grad_norm": 0.008739697746932507,
      "learning_rate": 1.5308523531543932e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 697
    },
    {
      "classification_loss": 0.0001,
      "epoch": 3.6980132450331125,
      "grad_norm": 0.0065392134711146355,
      "learning_rate": 1.5248370003250583e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 698
    },
    {
      "classification_loss": 0.0001,
      "epoch": 3.703311258278146,
      "grad_norm": 0.00461226049810648,
      "learning_rate": 1.5188261995761799e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 699
    },
    {
      "classification_loss": 0.0001,
      "epoch": 3.7086092715231787,
      "grad_norm": 0.004728569183498621,
      "learning_rate": 1.5128200084914692e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 700
    },
    {
      "epoch": 3.7086092715231787,
      "eval_accuracy": 0.9126760363578796,
      "eval_auroc": 0.9682399034500122,
      "eval_classification_loss": 0.6110203862190247,
      "eval_f1": 0.909621000289917,
      "eval_lm_loss": 0.05817615985870361,
      "eval_loss": 0.6691964864730835,
      "eval_precision": 0.9176470637321472,
      "eval_recall": 0.9017341136932373,
      "eval_runtime": 61.3535,
      "eval_samples_per_second": 5.786,
      "eval_steps_per_second": 2.901,
      "step": 700
    },
    {
      "classification_loss": 0.0001,
      "epoch": 3.713907284768212,
      "grad_norm": 0.012703921645879745,
      "learning_rate": 1.5068184846104764e-05,
      "lm_loss": 0.0,
      "loss": 0.0002,
      "step": 701
    },
    {
      "classification_loss": 0.0009,
      "epoch": 3.719205298013245,
      "grad_norm": 0.3630529046058655,
      "learning_rate": 1.5008216854280382e-05,
      "lm_loss": 0.0001,
      "loss": 0.001,
      "step": 702
    },
    {
      "classification_loss": 0.0002,
      "epoch": 3.724503311258278,
      "grad_norm": 0.010861873626708984,
      "learning_rate": 1.4948296683937305e-05,
      "lm_loss": 0.0,
      "loss": 0.0002,
      "step": 703
    },
    {
      "classification_loss": 0.0002,
      "epoch": 3.7298013245033115,
      "grad_norm": 0.020853715017437935,
      "learning_rate": 1.4888424909113139e-05,
      "lm_loss": 0.0,
      "loss": 0.0002,
      "step": 704
    },
    {
      "classification_loss": 0.0001,
      "epoch": 3.7350993377483444,
      "grad_norm": 0.003212082665413618,
      "learning_rate": 1.4828602103381889e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 705
    },
    {
      "classification_loss": 0.0001,
      "epoch": 3.7403973509933772,
      "grad_norm": 0.012315627187490463,
      "learning_rate": 1.476882883984839e-05,
      "lm_loss": 0.0,
      "loss": 0.0002,
      "step": 706
    },
    {
      "classification_loss": 0.0001,
      "epoch": 3.7456953642384105,
      "grad_norm": 0.020851658657193184,
      "learning_rate": 1.4709105691142915e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 707
    },
    {
      "classification_loss": 0.0002,
      "epoch": 3.750993377483444,
      "grad_norm": 0.024113405495882034,
      "learning_rate": 1.4649433229415588e-05,
      "lm_loss": 0.0,
      "loss": 0.0003,
      "step": 708
    },
    {
      "classification_loss": 0.0001,
      "epoch": 3.7562913907284767,
      "grad_norm": 0.009513629600405693,
      "learning_rate": 1.458981202633097e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 709
    },
    {
      "classification_loss": 0.0,
      "epoch": 3.76158940397351,
      "grad_norm": 0.0038411105051636696,
      "learning_rate": 1.4530242653062572e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 710
    },
    {
      "classification_loss": 0.0001,
      "epoch": 3.766887417218543,
      "grad_norm": 0.004456144291907549,
      "learning_rate": 1.4470725680287346e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 711
    },
    {
      "classification_loss": 0.0001,
      "epoch": 3.7721854304635762,
      "grad_norm": 0.008489495143294334,
      "learning_rate": 1.4411261678180277e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 712
    },
    {
      "classification_loss": 0.0002,
      "epoch": 3.777483443708609,
      "grad_norm": 0.14903484284877777,
      "learning_rate": 1.4351851216408856e-05,
      "lm_loss": 0.0,
      "loss": 0.0003,
      "step": 713
    },
    {
      "classification_loss": 0.0008,
      "epoch": 3.7827814569536424,
      "grad_norm": 0.35802167654037476,
      "learning_rate": 1.4292494864127693e-05,
      "lm_loss": 0.0001,
      "loss": 0.0009,
      "step": 714
    },
    {
      "classification_loss": 0.0001,
      "epoch": 3.7880794701986753,
      "grad_norm": 0.01702999323606491,
      "learning_rate": 1.4233193189972987e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 715
    },
    {
      "classification_loss": 0.0,
      "epoch": 3.7933774834437086,
      "grad_norm": 0.0024334643967449665,
      "learning_rate": 1.4173946762057144e-05,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 716
    },
    {
      "classification_loss": 0.0001,
      "epoch": 3.798675496688742,
      "grad_norm": 0.006429500412195921,
      "learning_rate": 1.411475614796331e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 717
    },
    {
      "classification_loss": 0.0001,
      "epoch": 3.803973509933775,
      "grad_norm": 0.006334204226732254,
      "learning_rate": 1.4055621914739915e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 718
    },
    {
      "classification_loss": 0.0001,
      "epoch": 3.8092715231788077,
      "grad_norm": 0.009716734290122986,
      "learning_rate": 1.3996544628895276e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 719
    },
    {
      "classification_loss": 0.0001,
      "epoch": 3.814569536423841,
      "grad_norm": 0.004858877509832382,
      "learning_rate": 1.3937524856392144e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 720
    },
    {
      "classification_loss": 0.0001,
      "epoch": 3.8198675496688743,
      "grad_norm": 0.004143928177654743,
      "learning_rate": 1.387856316264227e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 721
    },
    {
      "classification_loss": 0.0001,
      "epoch": 3.825165562913907,
      "grad_norm": 0.0057456959038972855,
      "learning_rate": 1.3819660112501054e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 722
    },
    {
      "classification_loss": 0.0001,
      "epoch": 3.8304635761589405,
      "grad_norm": 0.0038669791538268328,
      "learning_rate": 1.3760816270262044e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 723
    },
    {
      "classification_loss": 0.0002,
      "epoch": 3.8357615894039734,
      "grad_norm": 0.015411488711833954,
      "learning_rate": 1.3702032199651606e-05,
      "lm_loss": 0.0,
      "loss": 0.0002,
      "step": 724
    },
    {
      "classification_loss": 0.4737,
      "epoch": 3.8410596026490067,
      "grad_norm": 6.898221969604492,
      "learning_rate": 1.3643308463823472e-05,
      "lm_loss": 0.0485,
      "loss": 0.5222,
      "step": 725
    },
    {
      "classification_loss": 0.0001,
      "epoch": 3.8463576158940396,
      "grad_norm": 0.004028912167996168,
      "learning_rate": 1.358464562535336e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 726
    },
    {
      "classification_loss": 0.0,
      "epoch": 3.851655629139073,
      "grad_norm": 0.0023821331560611725,
      "learning_rate": 1.3526044246233609e-05,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 727
    },
    {
      "classification_loss": 0.0002,
      "epoch": 3.8569536423841058,
      "grad_norm": 0.010447257198393345,
      "learning_rate": 1.3467504887867757e-05,
      "lm_loss": 0.0,
      "loss": 0.0002,
      "step": 728
    },
    {
      "classification_loss": 0.0,
      "epoch": 3.862251655629139,
      "grad_norm": 0.002394022885710001,
      "learning_rate": 1.3409028111065194e-05,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 729
    },
    {
      "classification_loss": 0.0001,
      "epoch": 3.8675496688741724,
      "grad_norm": 0.005936316214501858,
      "learning_rate": 1.3350614476035775e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 730
    },
    {
      "classification_loss": 0.0001,
      "epoch": 3.8728476821192053,
      "grad_norm": 0.01174029242247343,
      "learning_rate": 1.3292264542384432e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 731
    },
    {
      "classification_loss": 0.0004,
      "epoch": 3.878145695364238,
      "grad_norm": 0.2955005466938019,
      "learning_rate": 1.3233978869105873e-05,
      "lm_loss": 0.0002,
      "loss": 0.0005,
      "step": 732
    },
    {
      "classification_loss": 0.0004,
      "epoch": 3.8834437086092715,
      "grad_norm": 0.07115139067173004,
      "learning_rate": 1.3175758014579152e-05,
      "lm_loss": 0.0001,
      "loss": 0.0005,
      "step": 733
    },
    {
      "classification_loss": 0.0001,
      "epoch": 3.888741721854305,
      "grad_norm": 0.009103034622967243,
      "learning_rate": 1.3117602536562389e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 734
    },
    {
      "classification_loss": 0.0001,
      "epoch": 3.8940397350993377,
      "grad_norm": 0.005303650163114071,
      "learning_rate": 1.3059512992187363e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 735
    },
    {
      "classification_loss": 0.0003,
      "epoch": 3.899337748344371,
      "grad_norm": 0.04587520658969879,
      "learning_rate": 1.3001489937954232e-05,
      "lm_loss": 0.0,
      "loss": 0.0004,
      "step": 736
    },
    {
      "classification_loss": 0.0002,
      "epoch": 3.904635761589404,
      "grad_norm": 0.012547608464956284,
      "learning_rate": 1.2943533929726171e-05,
      "lm_loss": 0.0,
      "loss": 0.0002,
      "step": 737
    },
    {
      "classification_loss": 0.0002,
      "epoch": 3.909933774834437,
      "grad_norm": 0.013149827718734741,
      "learning_rate": 1.288564552272403e-05,
      "lm_loss": 0.0,
      "loss": 0.0002,
      "step": 738
    },
    {
      "classification_loss": 0.0003,
      "epoch": 3.91523178807947,
      "grad_norm": 0.021581949666142464,
      "learning_rate": 1.2827825271521073e-05,
      "lm_loss": 0.0,
      "loss": 0.0003,
      "step": 739
    },
    {
      "classification_loss": 0.0001,
      "epoch": 3.9205298013245033,
      "grad_norm": 0.0046670641750097275,
      "learning_rate": 1.2770073730037588e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 740
    },
    {
      "classification_loss": 0.0001,
      "epoch": 3.9258278145695362,
      "grad_norm": 0.018754886463284492,
      "learning_rate": 1.2712391451535664e-05,
      "lm_loss": 0.0,
      "loss": 0.0002,
      "step": 741
    },
    {
      "classification_loss": 0.0002,
      "epoch": 3.9311258278145695,
      "grad_norm": 0.010284862481057644,
      "learning_rate": 1.2654778988613807e-05,
      "lm_loss": 0.0,
      "loss": 0.0002,
      "step": 742
    },
    {
      "classification_loss": 0.0001,
      "epoch": 3.936423841059603,
      "grad_norm": 0.007675246801227331,
      "learning_rate": 1.2597236893201712e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 743
    },
    {
      "classification_loss": 0.0001,
      "epoch": 3.9417218543046357,
      "grad_norm": 0.009307959116995335,
      "learning_rate": 1.2539765716554953e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 744
    },
    {
      "classification_loss": 0.0001,
      "epoch": 3.9470198675496686,
      "grad_norm": 0.011246738955378532,
      "learning_rate": 1.2482366009249676e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 745
    },
    {
      "classification_loss": 0.0001,
      "epoch": 3.952317880794702,
      "grad_norm": 0.006371213588863611,
      "learning_rate": 1.2425038321177376e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 746
    },
    {
      "classification_loss": 0.0002,
      "epoch": 3.9576158940397352,
      "grad_norm": 0.013936529867351055,
      "learning_rate": 1.2367783201539583e-05,
      "lm_loss": 0.0,
      "loss": 0.0002,
      "step": 747
    },
    {
      "classification_loss": 0.0001,
      "epoch": 3.962913907284768,
      "grad_norm": 0.008051518350839615,
      "learning_rate": 1.231060119884261e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 748
    },
    {
      "classification_loss": 0.0002,
      "epoch": 3.9682119205298014,
      "grad_norm": 0.013517788611352444,
      "learning_rate": 1.225349286089234e-05,
      "lm_loss": 0.0,
      "loss": 0.0002,
      "step": 749
    },
    {
      "classification_loss": 0.0001,
      "epoch": 3.9735099337748343,
      "grad_norm": 0.005143061280250549,
      "learning_rate": 1.219645873478891e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 750
    },
    {
      "epoch": 3.9735099337748343,
      "eval_accuracy": 0.9042253494262695,
      "eval_auroc": 0.9691768288612366,
      "eval_classification_loss": 0.6118075251579285,
      "eval_f1": 0.9017341136932373,
      "eval_lm_loss": 0.05866789072751999,
      "eval_loss": 0.670475423336029,
      "eval_precision": 0.9017341136932373,
      "eval_recall": 0.9017341136932373,
      "eval_runtime": 61.8018,
      "eval_samples_per_second": 5.744,
      "eval_steps_per_second": 2.88,
      "step": 750
    },
    {
      "classification_loss": 0.0805,
      "epoch": 3.9788079470198676,
      "grad_norm": 166.34323120117188,
      "learning_rate": 1.213949936692153e-05,
      "lm_loss": 0.0064,
      "loss": 0.0869,
      "step": 751
    },
    {
      "classification_loss": 0.0002,
      "epoch": 3.984105960264901,
      "grad_norm": 0.01689130626618862,
      "learning_rate": 1.2082615302963202e-05,
      "lm_loss": 0.0,
      "loss": 0.0002,
      "step": 752
    },
    {
      "classification_loss": 0.0001,
      "epoch": 3.989403973509934,
      "grad_norm": 0.009264855645596981,
      "learning_rate": 1.2025807087865533e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 753
    },
    {
      "classification_loss": 0.0002,
      "epoch": 3.9947019867549667,
      "grad_norm": 0.011881700716912746,
      "learning_rate": 1.1969075265853483e-05,
      "lm_loss": 0.0,
      "loss": 0.0002,
      "step": 754
    },
    {
      "classification_loss": 0.009,
      "epoch": 4.0,
      "grad_norm": 4.847745418548584,
      "learning_rate": 1.1912420380420155e-05,
      "lm_loss": 0.0031,
      "loss": 0.0121,
      "step": 755
    },
    {
      "classification_loss": 0.0001,
      "epoch": 4.005298013245033,
      "grad_norm": 0.00863186176866293,
      "learning_rate": 1.185584297432162e-05,
      "lm_loss": 0.0,
      "loss": 0.0002,
      "step": 756
    },
    {
      "classification_loss": 0.0001,
      "epoch": 4.010596026490067,
      "grad_norm": 0.004707596264779568,
      "learning_rate": 1.1799343589571679e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 757
    },
    {
      "classification_loss": 0.0002,
      "epoch": 4.015894039735099,
      "grad_norm": 0.013719013892114162,
      "learning_rate": 1.1742922767436688e-05,
      "lm_loss": 0.0,
      "loss": 0.0002,
      "step": 758
    },
    {
      "classification_loss": 0.0494,
      "epoch": 4.021192052980132,
      "grad_norm": 10.62270450592041,
      "learning_rate": 1.168658104843037e-05,
      "lm_loss": 0.0046,
      "loss": 0.0539,
      "step": 759
    },
    {
      "classification_loss": 0.0002,
      "epoch": 4.026490066225166,
      "grad_norm": 0.011772161349654198,
      "learning_rate": 1.163031897230863e-05,
      "lm_loss": 0.0,
      "loss": 0.0002,
      "step": 760
    },
    {
      "classification_loss": 0.0003,
      "epoch": 4.031788079470199,
      "grad_norm": 0.025081878527998924,
      "learning_rate": 1.1574137078064397e-05,
      "lm_loss": 0.0,
      "loss": 0.0003,
      "step": 761
    },
    {
      "classification_loss": 0.0001,
      "epoch": 4.0370860927152314,
      "grad_norm": 0.007727042771875858,
      "learning_rate": 1.151803590392246e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 762
    },
    {
      "classification_loss": 0.0001,
      "epoch": 4.042384105960265,
      "grad_norm": 0.009142504073679447,
      "learning_rate": 1.146201598733431e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 763
    },
    {
      "classification_loss": 0.0001,
      "epoch": 4.047682119205298,
      "grad_norm": 0.009661126881837845,
      "learning_rate": 1.1406077864972969e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 764
    },
    {
      "classification_loss": 0.0002,
      "epoch": 4.052980132450331,
      "grad_norm": 0.013046108186244965,
      "learning_rate": 1.135022207272787e-05,
      "lm_loss": 0.0,
      "loss": 0.0002,
      "step": 765
    },
    {
      "classification_loss": 0.0001,
      "epoch": 4.058278145695364,
      "grad_norm": 0.003973409999161959,
      "learning_rate": 1.129444914569974e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 766
    },
    {
      "classification_loss": 0.0001,
      "epoch": 4.063576158940397,
      "grad_norm": 0.004203333985060453,
      "learning_rate": 1.1238759618195427e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 767
    },
    {
      "classification_loss": 0.0003,
      "epoch": 4.0688741721854305,
      "grad_norm": 0.021326029673218727,
      "learning_rate": 1.1183154023722839e-05,
      "lm_loss": 0.0,
      "loss": 0.0003,
      "step": 768
    },
    {
      "classification_loss": 0.0001,
      "epoch": 4.074172185430464,
      "grad_norm": 0.0074068279936909676,
      "learning_rate": 1.1127632894985765e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 769
    },
    {
      "classification_loss": 0.0001,
      "epoch": 4.079470198675497,
      "grad_norm": 0.00855479296296835,
      "learning_rate": 1.1072196763878852e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 770
    },
    {
      "classification_loss": 0.0001,
      "epoch": 4.0847682119205295,
      "grad_norm": 0.005654282867908478,
      "learning_rate": 1.1016846161482427e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 771
    },
    {
      "classification_loss": 0.0001,
      "epoch": 4.090066225165563,
      "grad_norm": 0.009327742271125317,
      "learning_rate": 1.0961581618057461e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 772
    },
    {
      "classification_loss": 0.0,
      "epoch": 4.095364238410596,
      "grad_norm": 0.003114166436716914,
      "learning_rate": 1.090640366304049e-05,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 773
    },
    {
      "classification_loss": 0.0002,
      "epoch": 4.1006622516556295,
      "grad_norm": 0.012065451592206955,
      "learning_rate": 1.0851312825038527e-05,
      "lm_loss": 0.0,
      "loss": 0.0002,
      "step": 774
    },
    {
      "classification_loss": 0.0001,
      "epoch": 4.105960264900662,
      "grad_norm": 0.009513575583696365,
      "learning_rate": 1.0796309631823978e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 775
    },
    {
      "classification_loss": 0.0001,
      "epoch": 4.111258278145695,
      "grad_norm": 0.009891482070088387,
      "learning_rate": 1.0741394610329636e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 776
    },
    {
      "classification_loss": 0.0002,
      "epoch": 4.1165562913907285,
      "grad_norm": 0.013665454462170601,
      "learning_rate": 1.0686568286643584e-05,
      "lm_loss": 0.0,
      "loss": 0.0002,
      "step": 777
    },
    {
      "classification_loss": 0.0001,
      "epoch": 4.121854304635762,
      "grad_norm": 0.007335582282394171,
      "learning_rate": 1.06318311860042e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 778
    },
    {
      "classification_loss": 0.0001,
      "epoch": 4.127152317880794,
      "grad_norm": 0.004945365712046623,
      "learning_rate": 1.0577183832795068e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 779
    },
    {
      "classification_loss": 0.0001,
      "epoch": 4.132450331125828,
      "grad_norm": 0.008687729947268963,
      "learning_rate": 1.0522626750540029e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 780
    },
    {
      "classification_loss": 0.0001,
      "epoch": 4.137748344370861,
      "grad_norm": 0.004560061730444431,
      "learning_rate": 1.0468160461898091e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 781
    },
    {
      "classification_loss": 0.0001,
      "epoch": 4.143046357615894,
      "grad_norm": 0.009139570407569408,
      "learning_rate": 1.0413785488658463e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 782
    },
    {
      "classification_loss": 0.0001,
      "epoch": 4.1483443708609276,
      "grad_norm": 0.004179717972874641,
      "learning_rate": 1.0359502351735568e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 783
    },
    {
      "classification_loss": 0.0001,
      "epoch": 4.15364238410596,
      "grad_norm": 0.010165014304220676,
      "learning_rate": 1.0305311571164002e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 784
    },
    {
      "classification_loss": 0.0001,
      "epoch": 4.158940397350993,
      "grad_norm": 0.008153769187629223,
      "learning_rate": 1.025121366609362e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 785
    },
    {
      "classification_loss": 0.0001,
      "epoch": 4.164238410596027,
      "grad_norm": 0.006707365624606609,
      "learning_rate": 1.0197209154784491e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 786
    },
    {
      "classification_loss": 0.0001,
      "epoch": 4.16953642384106,
      "grad_norm": 0.009667054750025272,
      "learning_rate": 1.0143298554601994e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 787
    },
    {
      "classification_loss": 0.0001,
      "epoch": 4.174834437086092,
      "grad_norm": 0.011235668323934078,
      "learning_rate": 1.0089482382011816e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 788
    },
    {
      "classification_loss": 0.0001,
      "epoch": 4.180132450331126,
      "grad_norm": 0.004277492873370647,
      "learning_rate": 1.0035761152575036e-05,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 789
    },
    {
      "classification_loss": 0.0001,
      "epoch": 4.185430463576159,
      "grad_norm": 0.006855476647615433,
      "learning_rate": 9.982135380943183e-06,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 790
    },
    {
      "classification_loss": 0.0001,
      "epoch": 4.190728476821192,
      "grad_norm": 0.0038032149896025658,
      "learning_rate": 9.928605580853268e-06,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 791
    },
    {
      "classification_loss": 0.0001,
      "epoch": 4.196026490066225,
      "grad_norm": 0.0038023905362933874,
      "learning_rate": 9.87517226512292e-06,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 792
    },
    {
      "classification_loss": 0.0002,
      "epoch": 4.201324503311258,
      "grad_norm": 0.013985436409711838,
      "learning_rate": 9.821835945645426e-06,
      "lm_loss": 0.0,
      "loss": 0.0002,
      "step": 793
    },
    {
      "classification_loss": 0.0001,
      "epoch": 4.206622516556291,
      "grad_norm": 0.006130493711680174,
      "learning_rate": 9.768597133384843e-06,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 794
    },
    {
      "classification_loss": 0.0001,
      "epoch": 4.211920529801325,
      "grad_norm": 0.009510617703199387,
      "learning_rate": 9.715456338371128e-06,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 795
    },
    {
      "classification_loss": 0.0001,
      "epoch": 4.217218543046358,
      "grad_norm": 0.0063710142858326435,
      "learning_rate": 9.662414069695195e-06,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 796
    },
    {
      "classification_loss": 0.0,
      "epoch": 4.22251655629139,
      "grad_norm": 0.0017456067726016045,
      "learning_rate": 9.609470835504109e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 797
    },
    {
      "classification_loss": 0.0001,
      "epoch": 4.227814569536424,
      "grad_norm": 0.003975250758230686,
      "learning_rate": 9.55662714299615e-06,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 798
    },
    {
      "classification_loss": 0.0001,
      "epoch": 4.233112582781457,
      "grad_norm": 0.006270438898354769,
      "learning_rate": 9.503883498415991e-06,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 799
    },
    {
      "classification_loss": 0.0001,
      "epoch": 4.23841059602649,
      "grad_norm": 0.006832348182797432,
      "learning_rate": 9.451240407049865e-06,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 800
    },
    {
      "epoch": 4.23841059602649,
      "eval_accuracy": 0.9154929518699646,
      "eval_auroc": 0.9711300134658813,
      "eval_classification_loss": 0.5983288884162903,
      "eval_f1": 0.9122806787490845,
      "eval_lm_loss": 0.05832262709736824,
      "eval_loss": 0.6566515564918518,
      "eval_precision": 0.9230769276618958,
      "eval_recall": 0.9017341136932373,
      "eval_runtime": 59.5161,
      "eval_samples_per_second": 5.965,
      "eval_steps_per_second": 2.991,
      "step": 800
    },
    {
      "classification_loss": 0.0001,
      "epoch": 4.243708609271523,
      "grad_norm": 0.005713809747248888,
      "learning_rate": 9.398698373220663e-06,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 801
    },
    {
      "classification_loss": 0.0,
      "epoch": 4.249006622516556,
      "grad_norm": 0.002595488680526614,
      "learning_rate": 9.346257900283174e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 802
    },
    {
      "classification_loss": 0.0001,
      "epoch": 4.2543046357615895,
      "grad_norm": 0.004595302511006594,
      "learning_rate": 9.29391949061922e-06,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 803
    },
    {
      "classification_loss": 0.0001,
      "epoch": 4.259602649006623,
      "grad_norm": 0.006241128779947758,
      "learning_rate": 9.241683645632836e-06,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 804
    },
    {
      "classification_loss": 0.0001,
      "epoch": 4.264900662251655,
      "grad_norm": 0.006894748192280531,
      "learning_rate": 9.189550865745514e-06,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 805
    },
    {
      "classification_loss": 0.0001,
      "epoch": 4.2701986754966885,
      "grad_norm": 0.007266005035489798,
      "learning_rate": 9.137521650391343e-06,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 806
    },
    {
      "classification_loss": 0.0001,
      "epoch": 4.275496688741722,
      "grad_norm": 0.010608615353703499,
      "learning_rate": 9.085596498012295e-06,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 807
    },
    {
      "classification_loss": 0.0001,
      "epoch": 4.280794701986755,
      "grad_norm": 0.02118784189224243,
      "learning_rate": 9.033775906053381e-06,
      "lm_loss": 0.0001,
      "loss": 0.0002,
      "step": 808
    },
    {
      "classification_loss": 0.0001,
      "epoch": 4.2860927152317885,
      "grad_norm": 0.004617880564182997,
      "learning_rate": 8.982060370957953e-06,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 809
    },
    {
      "classification_loss": 0.0001,
      "epoch": 4.291390728476821,
      "grad_norm": 0.004088975489139557,
      "learning_rate": 8.930450388162884e-06,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 810
    },
    {
      "classification_loss": 0.0,
      "epoch": 4.296688741721854,
      "grad_norm": 0.0024079305585473776,
      "learning_rate": 8.878946452093862e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 811
    },
    {
      "classification_loss": 0.0,
      "epoch": 4.3019867549668875,
      "grad_norm": 0.0020385906100273132,
      "learning_rate": 8.827549056160657e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 812
    },
    {
      "classification_loss": 0.0,
      "epoch": 4.307284768211921,
      "grad_norm": 0.002797034103423357,
      "learning_rate": 8.776258692752355e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 813
    },
    {
      "classification_loss": 0.0,
      "epoch": 4.312582781456953,
      "grad_norm": 0.002882468281313777,
      "learning_rate": 8.725075853232694e-06,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 814
    },
    {
      "classification_loss": 0.0001,
      "epoch": 4.317880794701987,
      "grad_norm": 0.004914564546197653,
      "learning_rate": 8.674001027935299e-06,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 815
    },
    {
      "classification_loss": 0.0,
      "epoch": 4.32317880794702,
      "grad_norm": 0.0024123131297528744,
      "learning_rate": 8.623034706159039e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 816
    },
    {
      "classification_loss": 0.0,
      "epoch": 4.328476821192053,
      "grad_norm": 0.003387102158740163,
      "learning_rate": 8.572177376163315e-06,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 817
    },
    {
      "classification_loss": 0.0001,
      "epoch": 4.3337748344370866,
      "grad_norm": 0.004031948279589415,
      "learning_rate": 8.521429525163353e-06,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 818
    },
    {
      "classification_loss": 0.0,
      "epoch": 4.339072847682119,
      "grad_norm": 0.0029032782185822725,
      "learning_rate": 8.470791639325609e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 819
    },
    {
      "classification_loss": 0.0,
      "epoch": 4.344370860927152,
      "grad_norm": 0.0023927276488393545,
      "learning_rate": 8.420264203763031e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 820
    },
    {
      "classification_loss": 0.0002,
      "epoch": 4.349668874172186,
      "grad_norm": 0.018722867593169212,
      "learning_rate": 8.36984770253046e-06,
      "lm_loss": 0.0,
      "loss": 0.0002,
      "step": 821
    },
    {
      "classification_loss": 0.0,
      "epoch": 4.354966887417219,
      "grad_norm": 0.0010896361200138927,
      "learning_rate": 8.319542618619998e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 822
    },
    {
      "classification_loss": 0.0001,
      "epoch": 4.360264900662251,
      "grad_norm": 0.027253473177552223,
      "learning_rate": 8.269349433956339e-06,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 823
    },
    {
      "classification_loss": 0.0,
      "epoch": 4.365562913907285,
      "grad_norm": 0.0031694509088993073,
      "learning_rate": 8.219268629392206e-06,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 824
    },
    {
      "classification_loss": 0.0001,
      "epoch": 4.370860927152318,
      "grad_norm": 0.00450955331325531,
      "learning_rate": 8.169300684703682e-06,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 825
    },
    {
      "classification_loss": 0.0001,
      "epoch": 4.376158940397351,
      "grad_norm": 0.006932112853974104,
      "learning_rate": 8.119446078585687e-06,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 826
    },
    {
      "classification_loss": 0.0001,
      "epoch": 4.381456953642384,
      "grad_norm": 0.00432584946975112,
      "learning_rate": 8.069705288647323e-06,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 827
    },
    {
      "classification_loss": 0.0001,
      "epoch": 4.386754966887417,
      "grad_norm": 0.004954036325216293,
      "learning_rate": 8.02007879140733e-06,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 828
    },
    {
      "classification_loss": 0.0,
      "epoch": 4.39205298013245,
      "grad_norm": 0.003954713232815266,
      "learning_rate": 7.970567062289533e-06,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 829
    },
    {
      "classification_loss": 0.0001,
      "epoch": 4.397350993377484,
      "grad_norm": 0.005719243548810482,
      "learning_rate": 7.92117057561828e-06,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 830
    },
    {
      "classification_loss": 0.0001,
      "epoch": 4.402649006622516,
      "grad_norm": 0.004363866522908211,
      "learning_rate": 7.871889804613864e-06,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 831
    },
    {
      "classification_loss": 0.0,
      "epoch": 4.407947019867549,
      "grad_norm": 0.0015674940077587962,
      "learning_rate": 7.822725221388048e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 832
    },
    {
      "classification_loss": 0.0,
      "epoch": 4.413245033112583,
      "grad_norm": 0.002039646729826927,
      "learning_rate": 7.773677296939484e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 833
    },
    {
      "classification_loss": 0.0,
      "epoch": 4.418543046357616,
      "grad_norm": 0.0018777796067297459,
      "learning_rate": 7.724746501149252e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 834
    },
    {
      "classification_loss": 0.0001,
      "epoch": 4.423841059602649,
      "grad_norm": 0.006897266022861004,
      "learning_rate": 7.67593330277631e-06,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 835
    },
    {
      "classification_loss": 0.0003,
      "epoch": 4.429139072847682,
      "grad_norm": 0.035998281091451645,
      "learning_rate": 7.627238169453062e-06,
      "lm_loss": 0.0,
      "loss": 0.0004,
      "step": 836
    },
    {
      "classification_loss": 0.0,
      "epoch": 4.434437086092715,
      "grad_norm": 0.002366074360907078,
      "learning_rate": 7.578661567680812e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 837
    },
    {
      "classification_loss": 0.0,
      "epoch": 4.4397350993377485,
      "grad_norm": 0.005553464405238628,
      "learning_rate": 7.530203962825331e-06,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 838
    },
    {
      "classification_loss": 0.0,
      "epoch": 4.445033112582782,
      "grad_norm": 0.0013869141694158316,
      "learning_rate": 7.481865819112422e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 839
    },
    {
      "classification_loss": 0.0001,
      "epoch": 4.450331125827814,
      "grad_norm": 0.0036547270137816668,
      "learning_rate": 7.4336475996234085e-06,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 840
    },
    {
      "classification_loss": 0.0001,
      "epoch": 4.4556291390728475,
      "grad_norm": 0.0044960337691009045,
      "learning_rate": 7.385549766290767e-06,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 841
    },
    {
      "classification_loss": 0.0,
      "epoch": 4.460927152317881,
      "grad_norm": 0.004571199882775545,
      "learning_rate": 7.337572779893647e-06,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 842
    },
    {
      "classification_loss": 0.0,
      "epoch": 4.466225165562914,
      "grad_norm": 0.0031805045437067747,
      "learning_rate": 7.289717100053497e-06,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 843
    },
    {
      "classification_loss": 0.0001,
      "epoch": 4.4715231788079475,
      "grad_norm": 0.004189867060631514,
      "learning_rate": 7.2419831852296245e-06,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 844
    },
    {
      "classification_loss": 0.0,
      "epoch": 4.47682119205298,
      "grad_norm": 0.0016941447975113988,
      "learning_rate": 7.194371492714838e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 845
    },
    {
      "classification_loss": 0.0,
      "epoch": 4.482119205298013,
      "grad_norm": 0.0010875354055315256,
      "learning_rate": 7.1468824786310585e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 846
    },
    {
      "classification_loss": 0.0001,
      "epoch": 4.4874172185430465,
      "grad_norm": 0.01080703642219305,
      "learning_rate": 7.099516597924916e-06,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 847
    },
    {
      "classification_loss": 0.0002,
      "epoch": 4.49271523178808,
      "grad_norm": 0.01624026708304882,
      "learning_rate": 7.052274304363449e-06,
      "lm_loss": 0.0,
      "loss": 0.0002,
      "step": 848
    },
    {
      "classification_loss": 0.0,
      "epoch": 4.498013245033112,
      "grad_norm": 0.0024279868230223656,
      "learning_rate": 7.0051560505297e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 849
    },
    {
      "classification_loss": 0.0001,
      "epoch": 4.503311258278146,
      "grad_norm": 0.003783128224313259,
      "learning_rate": 6.958162287818415e-06,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 850
    },
    {
      "epoch": 4.503311258278146,
      "eval_accuracy": 0.9154929518699646,
      "eval_auroc": 0.971225380897522,
      "eval_classification_loss": 0.6189525127410889,
      "eval_f1": 0.9122806787490845,
      "eval_lm_loss": 0.05973050743341446,
      "eval_loss": 0.6786829829216003,
      "eval_precision": 0.9230769276618958,
      "eval_recall": 0.9017341136932373,
      "eval_runtime": 59.5024,
      "eval_samples_per_second": 5.966,
      "eval_steps_per_second": 2.991,
      "step": 850
    },
    {
      "classification_loss": 0.0,
      "epoch": 4.508609271523179,
      "grad_norm": 0.0056032598949968815,
      "learning_rate": 6.911293466431723e-06,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 851
    },
    {
      "classification_loss": 0.0001,
      "epoch": 4.513907284768212,
      "grad_norm": 0.014575241133570671,
      "learning_rate": 6.864550035374784e-06,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 852
    },
    {
      "classification_loss": 0.0001,
      "epoch": 4.519205298013245,
      "grad_norm": 0.004879848100244999,
      "learning_rate": 6.817932442451547e-06,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 853
    },
    {
      "classification_loss": 0.0,
      "epoch": 4.524503311258278,
      "grad_norm": 0.0038266675546765327,
      "learning_rate": 6.771441134260397e-06,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 854
    },
    {
      "classification_loss": 0.0,
      "epoch": 4.529801324503311,
      "grad_norm": 0.004485384561121464,
      "learning_rate": 6.725076556189915e-06,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 855
    },
    {
      "classification_loss": 0.0001,
      "epoch": 4.535099337748345,
      "grad_norm": 0.011933461762964725,
      "learning_rate": 6.678839152414616e-06,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 856
    },
    {
      "classification_loss": 0.0001,
      "epoch": 4.540397350993377,
      "grad_norm": 0.007339642383158207,
      "learning_rate": 6.632729365890653e-06,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 857
    },
    {
      "classification_loss": 0.0,
      "epoch": 4.54569536423841,
      "grad_norm": 0.0019227013690397143,
      "learning_rate": 6.5867476383516315e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 858
    },
    {
      "classification_loss": 0.0,
      "epoch": 4.550993377483444,
      "grad_norm": 0.002276953076943755,
      "learning_rate": 6.540894410304332e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 859
    },
    {
      "classification_loss": 0.0001,
      "epoch": 4.556291390728477,
      "grad_norm": 0.007578106597065926,
      "learning_rate": 6.495170121024494e-06,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 860
    },
    {
      "classification_loss": 0.0005,
      "epoch": 4.56158940397351,
      "grad_norm": 0.06761394441127777,
      "learning_rate": 6.449575208552643e-06,
      "lm_loss": 0.0001,
      "loss": 0.0006,
      "step": 861
    },
    {
      "classification_loss": 0.0001,
      "epoch": 4.566887417218543,
      "grad_norm": 0.009149789810180664,
      "learning_rate": 6.404110109689843e-06,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 862
    },
    {
      "classification_loss": 0.0,
      "epoch": 4.572185430463576,
      "grad_norm": 0.001154523342847824,
      "learning_rate": 6.358775259993564e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 863
    },
    {
      "classification_loss": 0.0002,
      "epoch": 4.577483443708609,
      "grad_norm": 0.017494484782218933,
      "learning_rate": 6.313571093773456e-06,
      "lm_loss": 0.0,
      "loss": 0.0002,
      "step": 864
    },
    {
      "classification_loss": 0.0,
      "epoch": 4.582781456953643,
      "grad_norm": 0.0016384042100980878,
      "learning_rate": 6.26849804408725e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 865
    },
    {
      "classification_loss": 0.0,
      "epoch": 4.588079470198675,
      "grad_norm": 0.0030941711738705635,
      "learning_rate": 6.223556542736544e-06,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 866
    },
    {
      "classification_loss": 0.0,
      "epoch": 4.593377483443708,
      "grad_norm": 0.0012507295468822122,
      "learning_rate": 6.178747020262708e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 867
    },
    {
      "classification_loss": 0.0,
      "epoch": 4.598675496688742,
      "grad_norm": 0.0033292812295258045,
      "learning_rate": 6.134069905942764e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 868
    },
    {
      "classification_loss": 0.0,
      "epoch": 4.603973509933775,
      "grad_norm": 0.0008498890092596412,
      "learning_rate": 6.089525627785231e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 869
    },
    {
      "classification_loss": 0.0001,
      "epoch": 4.609271523178808,
      "grad_norm": 0.02602836862206459,
      "learning_rate": 6.045114612526084e-06,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 870
    },
    {
      "classification_loss": 0.0001,
      "epoch": 4.614569536423841,
      "grad_norm": 0.008404335007071495,
      "learning_rate": 6.000837285624601e-06,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 871
    },
    {
      "classification_loss": 0.0,
      "epoch": 4.619867549668874,
      "grad_norm": 0.0034780220594257116,
      "learning_rate": 5.956694071259343e-06,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 872
    },
    {
      "classification_loss": 0.0,
      "epoch": 4.6251655629139075,
      "grad_norm": 0.0023306196089833975,
      "learning_rate": 5.912685392324074e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 873
    },
    {
      "classification_loss": 0.0001,
      "epoch": 4.630463576158941,
      "grad_norm": 0.01609618403017521,
      "learning_rate": 5.868811670423669e-06,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 874
    },
    {
      "classification_loss": 0.0,
      "epoch": 4.635761589403973,
      "grad_norm": 0.0022946821991354227,
      "learning_rate": 5.825073325870149e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 875
    },
    {
      "classification_loss": 0.0001,
      "epoch": 4.6410596026490065,
      "grad_norm": 0.009495827369391918,
      "learning_rate": 5.781470777678591e-06,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 876
    },
    {
      "classification_loss": 0.0,
      "epoch": 4.64635761589404,
      "grad_norm": 0.003856204217299819,
      "learning_rate": 5.738004443563132e-06,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 877
    },
    {
      "classification_loss": 0.0,
      "epoch": 4.651655629139073,
      "grad_norm": 0.0017502924893051386,
      "learning_rate": 5.694674739932999e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 878
    },
    {
      "classification_loss": 0.0,
      "epoch": 4.656953642384106,
      "grad_norm": 0.002894430421292782,
      "learning_rate": 5.6514820818884644e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 879
    },
    {
      "classification_loss": 0.0001,
      "epoch": 4.662251655629139,
      "grad_norm": 0.0051035210490226746,
      "learning_rate": 5.608426883216927e-06,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 880
    },
    {
      "classification_loss": 0.0,
      "epoch": 4.667549668874172,
      "grad_norm": 0.0048748948611319065,
      "learning_rate": 5.565509556388889e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 881
    },
    {
      "classification_loss": 0.0001,
      "epoch": 4.6728476821192055,
      "grad_norm": 0.016885599121451378,
      "learning_rate": 5.522730512554064e-06,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 882
    },
    {
      "classification_loss": 0.0,
      "epoch": 4.678145695364238,
      "grad_norm": 0.0010629016906023026,
      "learning_rate": 5.480090161537388e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 883
    },
    {
      "classification_loss": 0.0,
      "epoch": 4.683443708609271,
      "grad_norm": 0.0031312399078160524,
      "learning_rate": 5.437588911835119e-06,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 884
    },
    {
      "classification_loss": 0.0,
      "epoch": 4.688741721854305,
      "grad_norm": 0.0010784793412312865,
      "learning_rate": 5.395227170610924e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 885
    },
    {
      "classification_loss": 0.0,
      "epoch": 4.694039735099338,
      "grad_norm": 0.0017198675777763128,
      "learning_rate": 5.353005343691972e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 886
    },
    {
      "classification_loss": 0.0,
      "epoch": 4.699337748344371,
      "grad_norm": 0.003037248505279422,
      "learning_rate": 5.310923835565052e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 887
    },
    {
      "classification_loss": 0.0,
      "epoch": 4.704635761589404,
      "grad_norm": 0.0023650797083973885,
      "learning_rate": 5.268983049372682e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 888
    },
    {
      "classification_loss": 0.0,
      "epoch": 4.709933774834437,
      "grad_norm": 0.0021326253190636635,
      "learning_rate": 5.227183386909258e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 889
    },
    {
      "classification_loss": 0.0,
      "epoch": 4.71523178807947,
      "grad_norm": 0.0023955833166837692,
      "learning_rate": 5.18552524861722e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 890
    },
    {
      "classification_loss": 0.0,
      "epoch": 4.720529801324504,
      "grad_norm": 0.005118196830153465,
      "learning_rate": 5.144009033583182e-06,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 891
    },
    {
      "classification_loss": 0.0,
      "epoch": 4.725827814569536,
      "grad_norm": 0.0017677193973213434,
      "learning_rate": 5.102635139534149e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 892
    },
    {
      "classification_loss": 0.0001,
      "epoch": 4.731125827814569,
      "grad_norm": 0.003691599005833268,
      "learning_rate": 5.061403962833669e-06,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 893
    },
    {
      "classification_loss": 0.0,
      "epoch": 4.736423841059603,
      "grad_norm": 0.004282246343791485,
      "learning_rate": 5.020315898478045e-06,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 894
    },
    {
      "classification_loss": 0.0,
      "epoch": 4.741721854304636,
      "grad_norm": 0.0020965307485312223,
      "learning_rate": 4.979371340092589e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 895
    },
    {
      "classification_loss": 0.0001,
      "epoch": 4.747019867549669,
      "grad_norm": 0.012031416408717632,
      "learning_rate": 4.938570679927784e-06,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 896
    },
    {
      "classification_loss": 0.0001,
      "epoch": 4.752317880794702,
      "grad_norm": 0.004634873941540718,
      "learning_rate": 4.897914308855589e-06,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 897
    },
    {
      "classification_loss": 0.0,
      "epoch": 4.757615894039735,
      "grad_norm": 0.0012931706150993705,
      "learning_rate": 4.857402616365652e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 898
    },
    {
      "classification_loss": 0.0,
      "epoch": 4.762913907284768,
      "grad_norm": 0.0018149279057979584,
      "learning_rate": 4.8170359905616024e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 899
    },
    {
      "classification_loss": 0.0,
      "epoch": 4.768211920529802,
      "grad_norm": 0.002615417819470167,
      "learning_rate": 4.7768148181573295e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 900
    },
    {
      "epoch": 4.768211920529802,
      "eval_accuracy": 0.9211267828941345,
      "eval_auroc": 0.9711935520172119,
      "eval_classification_loss": 0.6371268033981323,
      "eval_f1": 0.9186046719551086,
      "eval_lm_loss": 0.060832321643829346,
      "eval_loss": 0.6979591250419617,
      "eval_precision": 0.9239766001701355,
      "eval_recall": 0.913294792175293,
      "eval_runtime": 59.3521,
      "eval_samples_per_second": 5.981,
      "eval_steps_per_second": 2.999,
      "step": 900
    },
    {
      "classification_loss": 0.0,
      "epoch": 4.773509933774834,
      "grad_norm": 0.0023941807448863983,
      "learning_rate": 4.736739484473252e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 901
    },
    {
      "classification_loss": 0.0,
      "epoch": 4.778807947019867,
      "grad_norm": 0.0011293271090835333,
      "learning_rate": 4.696810373432679e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 902
    },
    {
      "classification_loss": 0.0,
      "epoch": 4.784105960264901,
      "grad_norm": 0.0031923505011945963,
      "learning_rate": 4.6570278675580685e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 903
    },
    {
      "classification_loss": 0.0001,
      "epoch": 4.789403973509934,
      "grad_norm": 0.013516956940293312,
      "learning_rate": 4.617392347967422e-06,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 904
    },
    {
      "classification_loss": 0.0,
      "epoch": 4.7947019867549665,
      "grad_norm": 0.0009872688679024577,
      "learning_rate": 4.577904194370595e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 905
    },
    {
      "classification_loss": 0.0,
      "epoch": 4.8,
      "grad_norm": 0.0016611311584711075,
      "learning_rate": 4.53856378506566e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 906
    },
    {
      "classification_loss": 0.0,
      "epoch": 4.805298013245033,
      "grad_norm": 0.0026201612781733274,
      "learning_rate": 4.499371496935319e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 907
    },
    {
      "classification_loss": 0.0,
      "epoch": 4.8105960264900665,
      "grad_norm": 0.002635297132655978,
      "learning_rate": 4.460327705443239e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 908
    },
    {
      "classification_loss": 0.0,
      "epoch": 4.815894039735099,
      "grad_norm": 0.0015646726824343204,
      "learning_rate": 4.42143278463051e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 909
    },
    {
      "classification_loss": 0.0001,
      "epoch": 4.821192052980132,
      "grad_norm": 0.003791561583057046,
      "learning_rate": 4.382687107112018e-06,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 910
    },
    {
      "classification_loss": 0.0,
      "epoch": 4.8264900662251655,
      "grad_norm": 0.0053890482522547245,
      "learning_rate": 4.344091044072889e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 911
    },
    {
      "classification_loss": 0.0,
      "epoch": 4.831788079470199,
      "grad_norm": 0.01122351922094822,
      "learning_rate": 4.305644965264948e-06,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 912
    },
    {
      "classification_loss": 0.0,
      "epoch": 4.837086092715232,
      "grad_norm": 0.0012640550266951323,
      "learning_rate": 4.267349239003169e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 913
    },
    {
      "classification_loss": 0.0001,
      "epoch": 4.842384105960265,
      "grad_norm": 0.004840104840695858,
      "learning_rate": 4.229204232162123e-06,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 914
    },
    {
      "classification_loss": 0.0,
      "epoch": 4.847682119205298,
      "grad_norm": 0.002192526590079069,
      "learning_rate": 4.191210310172504e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 915
    },
    {
      "classification_loss": 0.0,
      "epoch": 4.852980132450331,
      "grad_norm": 0.0010526332771405578,
      "learning_rate": 4.153367837017588e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 916
    },
    {
      "classification_loss": 0.0,
      "epoch": 4.8582781456953645,
      "grad_norm": 0.0007292481022886932,
      "learning_rate": 4.115677175229784e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 917
    },
    {
      "classification_loss": 0.0,
      "epoch": 4.863576158940397,
      "grad_norm": 0.0010620016837492585,
      "learning_rate": 4.078138685887125e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 918
    },
    {
      "classification_loss": 0.0,
      "epoch": 4.86887417218543,
      "grad_norm": 0.0022890870459377766,
      "learning_rate": 4.04075272860984e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 919
    },
    {
      "classification_loss": 0.0,
      "epoch": 4.874172185430464,
      "grad_norm": 0.0035543786361813545,
      "learning_rate": 4.003519661556881e-06,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 920
    },
    {
      "classification_loss": 0.0,
      "epoch": 4.879470198675497,
      "grad_norm": 0.00301356241106987,
      "learning_rate": 3.966439841422523e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 921
    },
    {
      "classification_loss": 0.0001,
      "epoch": 4.88476821192053,
      "grad_norm": 0.0038982063997536898,
      "learning_rate": 3.92951362343291e-06,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 922
    },
    {
      "classification_loss": 0.0,
      "epoch": 4.890066225165563,
      "grad_norm": 0.0027123596519231796,
      "learning_rate": 3.892741361342682e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 923
    },
    {
      "classification_loss": 0.0001,
      "epoch": 4.895364238410596,
      "grad_norm": 0.0048912763595581055,
      "learning_rate": 3.85612340743158e-06,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 924
    },
    {
      "classification_loss": 0.0,
      "epoch": 4.900662251655629,
      "grad_norm": 0.002730341861024499,
      "learning_rate": 3.819660112501053e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 925
    },
    {
      "classification_loss": 0.0,
      "epoch": 4.905960264900663,
      "grad_norm": 0.002491772873327136,
      "learning_rate": 3.7833518258709266e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 926
    },
    {
      "classification_loss": 0.0,
      "epoch": 4.911258278145695,
      "grad_norm": 0.002166047226637602,
      "learning_rate": 3.747198895376023e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 927
    },
    {
      "classification_loss": 0.0,
      "epoch": 4.916556291390728,
      "grad_norm": 0.00207554385997355,
      "learning_rate": 3.7112016673628604e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 928
    },
    {
      "classification_loss": 0.0,
      "epoch": 4.921854304635762,
      "grad_norm": 0.0011283083586022258,
      "learning_rate": 3.675360486686321e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 929
    },
    {
      "classification_loss": 0.0,
      "epoch": 4.927152317880795,
      "grad_norm": 0.001195248099975288,
      "learning_rate": 3.6396756967063353e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 930
    },
    {
      "classification_loss": 0.0,
      "epoch": 4.932450331125828,
      "grad_norm": 0.0015885410830378532,
      "learning_rate": 3.604147639284619e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 931
    },
    {
      "classification_loss": 0.0,
      "epoch": 4.937748344370861,
      "grad_norm": 0.002894699340686202,
      "learning_rate": 3.568776654781374e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 932
    },
    {
      "classification_loss": 0.0,
      "epoch": 4.943046357615894,
      "grad_norm": 0.0010517064947634935,
      "learning_rate": 3.5335630820520363e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 933
    },
    {
      "classification_loss": 0.0,
      "epoch": 4.948344370860927,
      "grad_norm": 0.0010577056091278791,
      "learning_rate": 3.4985072584440417e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 934
    },
    {
      "classification_loss": 0.0,
      "epoch": 4.95364238410596,
      "grad_norm": 0.0022542839869856834,
      "learning_rate": 3.463609519793567e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 935
    },
    {
      "classification_loss": 0.0001,
      "epoch": 4.958940397350993,
      "grad_norm": 0.022400394082069397,
      "learning_rate": 3.4288702004223495e-06,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 936
    },
    {
      "classification_loss": 0.0,
      "epoch": 4.964238410596026,
      "grad_norm": 0.006658296566456556,
      "learning_rate": 3.394289633134442e-06,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 937
    },
    {
      "classification_loss": 0.0,
      "epoch": 4.96953642384106,
      "grad_norm": 0.002040076768025756,
      "learning_rate": 3.359868149213066e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 938
    },
    {
      "classification_loss": 0.0001,
      "epoch": 4.974834437086093,
      "grad_norm": 0.007481524720788002,
      "learning_rate": 3.3256060784174003e-06,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 939
    },
    {
      "classification_loss": 0.0,
      "epoch": 4.9801324503311255,
      "grad_norm": 0.0019526060204952955,
      "learning_rate": 3.2915037489794456e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 940
    },
    {
      "classification_loss": 0.0,
      "epoch": 4.985430463576159,
      "grad_norm": 0.0015010767383500934,
      "learning_rate": 3.2575614876008798e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 941
    },
    {
      "classification_loss": 0.0,
      "epoch": 4.990728476821192,
      "grad_norm": 0.0008530152845196426,
      "learning_rate": 3.2237796194499204e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 942
    },
    {
      "classification_loss": 0.0,
      "epoch": 4.9960264900662255,
      "grad_norm": 0.004138030111789703,
      "learning_rate": 3.190158468158209e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 943
    },
    {
      "classification_loss": 0.0001,
      "epoch": 5.001324503311258,
      "grad_norm": 0.00470867520198226,
      "learning_rate": 3.156698355817713e-06,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 944
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.006622516556291,
      "grad_norm": 0.0008700286853127182,
      "learning_rate": 3.123399602977637e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 945
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.0119205298013245,
      "grad_norm": 0.00248768855817616,
      "learning_rate": 3.0902625286413656e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 946
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.017218543046358,
      "grad_norm": 0.002072117058560252,
      "learning_rate": 3.057287450263382e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 947
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.022516556291391,
      "grad_norm": 0.0019320343853905797,
      "learning_rate": 3.024474683746261e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 948
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.027814569536424,
      "grad_norm": 0.0009245191467925906,
      "learning_rate": 2.991824543437607e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 949
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.033112582781457,
      "grad_norm": 0.0018495934782549739,
      "learning_rate": 2.959337342127058e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 950
    },
    {
      "epoch": 5.033112582781457,
      "eval_accuracy": 0.9183098673820496,
      "eval_auroc": 0.9707329869270325,
      "eval_classification_loss": 0.6499205827713013,
      "eval_f1": 0.9154518842697144,
      "eval_lm_loss": 0.06177937611937523,
      "eval_loss": 0.711699903011322,
      "eval_precision": 0.9235293865203857,
      "eval_recall": 0.9075144529342651,
      "eval_runtime": 59.6377,
      "eval_samples_per_second": 5.953,
      "eval_steps_per_second": 2.985,
      "step": 950
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.03841059602649,
      "grad_norm": 0.0015338525408878922,
      "learning_rate": 2.927013391043305e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 951
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.0437086092715235,
      "grad_norm": 0.0052425554022192955,
      "learning_rate": 2.8948529998510787e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 952
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.049006622516556,
      "grad_norm": 0.0009584020008333027,
      "learning_rate": 2.8628564766482193e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 953
    },
    {
      "classification_loss": 0.0001,
      "epoch": 5.054304635761589,
      "grad_norm": 0.013620393350720406,
      "learning_rate": 2.8310241279626784e-06,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 954
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.059602649006623,
      "grad_norm": 0.002365581225603819,
      "learning_rate": 2.7993562587496324e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 955
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.064900662251656,
      "grad_norm": 0.0018494047690182924,
      "learning_rate": 2.7678531723885325e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 956
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.070198675496688,
      "grad_norm": 0.0010272490326315165,
      "learning_rate": 2.7365151706801918e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 957
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.075496688741722,
      "grad_norm": 0.0013477844186127186,
      "learning_rate": 2.7053425538439238e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 958
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.080794701986755,
      "grad_norm": 0.0013469436671584845,
      "learning_rate": 2.6743356205146276e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 959
    },
    {
      "classification_loss": 0.0001,
      "epoch": 5.086092715231788,
      "grad_norm": 0.008101066574454308,
      "learning_rate": 2.643494667739972e-06,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 960
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.091390728476822,
      "grad_norm": 0.0027826528530567884,
      "learning_rate": 2.6128199909775066e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 961
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.096688741721854,
      "grad_norm": 0.0013325174804776907,
      "learning_rate": 2.582311884091848e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 962
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.101986754966887,
      "grad_norm": 0.0026630694046616554,
      "learning_rate": 2.5519706393518907e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 963
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.107284768211921,
      "grad_norm": 0.0025108542758971453,
      "learning_rate": 2.521796547427953e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 964
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.112582781456954,
      "grad_norm": 0.0018504391191527247,
      "learning_rate": 2.49178989738905e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 965
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.117880794701986,
      "grad_norm": 0.0037515030708163977,
      "learning_rate": 2.461950976700074e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 966
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.12317880794702,
      "grad_norm": 0.0013631234178319573,
      "learning_rate": 2.432280071219075e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 967
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.128476821192053,
      "grad_norm": 0.0021692269947379827,
      "learning_rate": 2.40277746519451e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 968
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.133774834437086,
      "grad_norm": 0.001952878199517727,
      "learning_rate": 2.373443441262524e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 969
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.139072847682119,
      "grad_norm": 0.0012436609249562025,
      "learning_rate": 2.3442782804442253e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 970
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.144370860927152,
      "grad_norm": 0.002197202295064926,
      "learning_rate": 2.315282262143028e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 971
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.149668874172185,
      "grad_norm": 0.0018387071322649717,
      "learning_rate": 2.286455664141929e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 972
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.154966887417219,
      "grad_norm": 0.0010894397273659706,
      "learning_rate": 2.2577987626008934e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 973
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.160264900662252,
      "grad_norm": 0.0012541967444121838,
      "learning_rate": 2.229311832054175e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 974
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.1655629139072845,
      "grad_norm": 0.0010804926278069615,
      "learning_rate": 2.2009951454077073e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 975
    },
    {
      "classification_loss": 0.0001,
      "epoch": 5.170860927152318,
      "grad_norm": 0.004684157203882933,
      "learning_rate": 2.17284897393647e-06,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 976
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.176158940397351,
      "grad_norm": 0.0017460857052356005,
      "learning_rate": 2.1448735872819125e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 977
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.1814569536423845,
      "grad_norm": 0.002378842793405056,
      "learning_rate": 2.1170692534493486e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 978
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.186754966887417,
      "grad_norm": 0.0033449253533035517,
      "learning_rate": 2.0894362388054023e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 979
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.19205298013245,
      "grad_norm": 0.0008330606506206095,
      "learning_rate": 2.0619748080754596e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 980
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.1973509933774835,
      "grad_norm": 0.000840048014651984,
      "learning_rate": 2.0346852243411154e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 981
    },
    {
      "classification_loss": 0.0001,
      "epoch": 5.202649006622517,
      "grad_norm": 0.004155600443482399,
      "learning_rate": 2.0075677490376687e-06,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 982
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.207947019867549,
      "grad_norm": 0.0010943934321403503,
      "learning_rate": 1.9806226419516195e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 983
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.213245033112583,
      "grad_norm": 0.0029910020530223846,
      "learning_rate": 1.953850161218156e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 984
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.218543046357616,
      "grad_norm": 0.0029405381064862013,
      "learning_rate": 1.9272505633187212e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 985
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.223841059602649,
      "grad_norm": 0.0019608563743531704,
      "learning_rate": 1.9008241030785112e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 986
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.2291390728476825,
      "grad_norm": 0.0036305475514382124,
      "learning_rate": 1.8745710336640743e-06,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 987
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.234437086092715,
      "grad_norm": 0.0008591851801611483,
      "learning_rate": 1.8484916065808622e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 988
    },
    {
      "classification_loss": 0.0001,
      "epoch": 5.239735099337748,
      "grad_norm": 0.00726384948939085,
      "learning_rate": 1.8225860716708155e-06,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 989
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.245033112582782,
      "grad_norm": 0.0018332048784941435,
      "learning_rate": 1.7968546771100005e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 990
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.250331125827815,
      "grad_norm": 0.003699413500726223,
      "learning_rate": 1.7712976694061934e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 991
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.255629139072847,
      "grad_norm": 0.0019689376931637526,
      "learning_rate": 1.745915293396554e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 992
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.260927152317881,
      "grad_norm": 0.0010295313550159335,
      "learning_rate": 1.7207077922452465e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 993
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.266225165562914,
      "grad_norm": 0.0029155125375837088,
      "learning_rate": 1.695675407441142e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 994
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.271523178807947,
      "grad_norm": 0.0032771106343716383,
      "learning_rate": 1.6708183787954801e-06,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 995
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.27682119205298,
      "grad_norm": 0.0018499705474823713,
      "learning_rate": 1.6461369444395891e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 996
    },
    {
      "classification_loss": 0.0001,
      "epoch": 5.282119205298013,
      "grad_norm": 0.01044908631592989,
      "learning_rate": 1.621631340822587e-06,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 997
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.287417218543046,
      "grad_norm": 0.00234129186719656,
      "learning_rate": 1.5973018027091346e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 998
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.29271523178808,
      "grad_norm": 0.0016064835945144296,
      "learning_rate": 1.5731485631771803e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 999
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.298013245033113,
      "grad_norm": 0.0015727599384263158,
      "learning_rate": 1.5491718536157163e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 1000
    },
    {
      "epoch": 5.298013245033113,
      "eval_accuracy": 0.9183098673820496,
      "eval_auroc": 0.9709553122520447,
      "eval_classification_loss": 0.6516768932342529,
      "eval_f1": 0.9154518842697144,
      "eval_lm_loss": 0.06193294748663902,
      "eval_loss": 0.7136098146438599,
      "eval_precision": 0.9235293865203857,
      "eval_recall": 0.9075144529342651,
      "eval_runtime": 59.2418,
      "eval_samples_per_second": 5.992,
      "eval_steps_per_second": 3.005,
      "step": 1000
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.303311258278145,
      "grad_norm": 0.005457419902086258,
      "learning_rate": 1.5253719037225745e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 1001
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.308609271523179,
      "grad_norm": 0.0008675518911331892,
      "learning_rate": 1.5017489415022235e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 1002
    },
    {
      "classification_loss": 0.0001,
      "epoch": 5.313907284768212,
      "grad_norm": 0.006338061764836311,
      "learning_rate": 1.4783031932635816e-06,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 1003
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.319205298013245,
      "grad_norm": 0.0009088020888157189,
      "learning_rate": 1.455034883617852e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 1004
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.324503311258278,
      "grad_norm": 0.0026775801088660955,
      "learning_rate": 1.431944235476368e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 1005
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.329801324503311,
      "grad_norm": 0.0013814802514389157,
      "learning_rate": 1.409031470048452e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 1006
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.335099337748344,
      "grad_norm": 0.0017103147692978382,
      "learning_rate": 1.386296806839318e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 1007
    },
    {
      "classification_loss": 0.0001,
      "epoch": 5.340397350993378,
      "grad_norm": 0.004955066367983818,
      "learning_rate": 1.363740463647938e-06,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 1008
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.34569536423841,
      "grad_norm": 0.0012238359777256846,
      "learning_rate": 1.341362656564984e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 1009
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.3509933774834435,
      "grad_norm": 0.0019284947775304317,
      "learning_rate": 1.319163599970732e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 1010
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.356291390728477,
      "grad_norm": 0.0016025822842493653,
      "learning_rate": 1.2971435065330295e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 1011
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.36158940397351,
      "grad_norm": 0.0008371240692213178,
      "learning_rate": 1.275302587205256e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 1012
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.3668874172185435,
      "grad_norm": 0.0029431029688566923,
      "learning_rate": 1.2536410512242792e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 1013
    },
    {
      "classification_loss": 0.0001,
      "epoch": 5.372185430463576,
      "grad_norm": 0.011103352531790733,
      "learning_rate": 1.2321591061084837e-06,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 1014
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.377483443708609,
      "grad_norm": 0.0014377802144736052,
      "learning_rate": 1.2108569576557527e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 1015
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.3827814569536425,
      "grad_norm": 0.0009138718596659601,
      "learning_rate": 1.1897348099415207e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 1016
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.388079470198676,
      "grad_norm": 0.0020581709686666727,
      "learning_rate": 1.1687928653168034e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 1017
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.393377483443708,
      "grad_norm": 0.0008572710794396698,
      "learning_rate": 1.1480313244062603e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 1018
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.398675496688742,
      "grad_norm": 0.0009664894314482808,
      "learning_rate": 1.1274503861062857e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 1019
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.403973509933775,
      "grad_norm": 0.0029451814480125904,
      "learning_rate": 1.1070502475830814e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 1020
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.409271523178808,
      "grad_norm": 0.0030417924281209707,
      "learning_rate": 1.0868311042707936e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 1021
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.414569536423841,
      "grad_norm": 0.0017645757179707289,
      "learning_rate": 1.0667931498696182e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 1022
    },
    {
      "classification_loss": 0.0001,
      "epoch": 5.419867549668874,
      "grad_norm": 0.017177866771817207,
      "learning_rate": 1.0469365763439532e-06,
      "lm_loss": 0.0001,
      "loss": 0.0002,
      "step": 1023
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.425165562913907,
      "grad_norm": 0.002458690432831645,
      "learning_rate": 1.0272615739205638e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 1024
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.430463576158941,
      "grad_norm": 0.0034395207185298204,
      "learning_rate": 1.0077683310867602e-06,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 1025
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.435761589403974,
      "grad_norm": 0.0008364359382539988,
      "learning_rate": 9.884570345885858e-07,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 1026
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.441059602649006,
      "grad_norm": 0.0011205412447452545,
      "learning_rate": 9.693278694290243e-07,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 1027
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.44635761589404,
      "grad_norm": 0.001807920983992517,
      "learning_rate": 9.503810188662465e-07,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 1028
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.451655629139073,
      "grad_norm": 0.0011727686505764723,
      "learning_rate": 9.316166644118319e-07,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 1029
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.456953642384106,
      "grad_norm": 0.0006516680005006492,
      "learning_rate": 9.130349858290377e-07,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 1030
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.462251655629139,
      "grad_norm": 0.0015733666950836778,
      "learning_rate": 8.946361611310949e-07,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 1031
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.467549668874172,
      "grad_norm": 0.002623822772875428,
      "learning_rate": 8.764203665794691e-07,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 1032
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.472847682119205,
      "grad_norm": 0.0010173511691391468,
      "learning_rate": 8.583877766822013e-07,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 1033
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.478145695364239,
      "grad_norm": 0.0009907081257551908,
      "learning_rate": 8.405385641922215e-07,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 1034
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.483443708609272,
      "grad_norm": 0.001547601306810975,
      "learning_rate": 8.228729001056912e-07,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 1035
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.488741721854304,
      "grad_norm": 0.0016221102559939027,
      "learning_rate": 8.053909536603788e-07,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 1036
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.494039735099338,
      "grad_norm": 0.0018374106148257852,
      "learning_rate": 7.88092892334027e-07,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 1037
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.499337748344371,
      "grad_norm": 0.0015565954381600022,
      "learning_rate": 7.709788818427522e-07,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 1038
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.5046357615894035,
      "grad_norm": 0.0009528156369924545,
      "learning_rate": 7.54049086139459e-07,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 1039
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.509933774834437,
      "grad_norm": 0.0010967126581817865,
      "learning_rate": 7.373036674122591e-07,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 1040
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.51523178807947,
      "grad_norm": 0.001046857563778758,
      "learning_rate": 7.207427860829352e-07,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 1041
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.520529801324503,
      "grad_norm": 0.001787831773981452,
      "learning_rate": 7.043666008053862e-07,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 1042
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.525827814569537,
      "grad_norm": 0.0006924797780811787,
      "learning_rate": 6.881752684641219e-07,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 1043
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.531125827814569,
      "grad_norm": 0.001525316620245576,
      "learning_rate": 6.72168944172753e-07,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 1044
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.5364238410596025,
      "grad_norm": 0.002380039542913437,
      "learning_rate": 6.56347781272495e-07,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 1045
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.541721854304636,
      "grad_norm": 0.0021851728670299053,
      "learning_rate": 6.407119313307286e-07,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 1046
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.547019867549669,
      "grad_norm": 0.000983169418759644,
      "learning_rate": 6.252615441395105e-07,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 1047
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.552317880794702,
      "grad_norm": 0.0012193166185170412,
      "learning_rate": 6.099967677141716e-07,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 1048
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.557615894039735,
      "grad_norm": 0.0008099583792500198,
      "learning_rate": 5.949177482918744e-07,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 1049
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.562913907284768,
      "grad_norm": 0.002816656371578574,
      "learning_rate": 5.800246303302271e-07,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 1050
    },
    {
      "epoch": 5.562913907284768,
      "eval_accuracy": 0.9183098673820496,
      "eval_auroc": 0.9711617827415466,
      "eval_classification_loss": 0.6492831110954285,
      "eval_f1": 0.9154518842697144,
      "eval_lm_loss": 0.061807069927453995,
      "eval_loss": 0.7110901474952698,
      "eval_precision": 0.9235293865203857,
      "eval_recall": 0.9075144529342651,
      "eval_runtime": 59.2245,
      "eval_samples_per_second": 5.994,
      "eval_steps_per_second": 3.006,
      "step": 1050
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.5682119205298015,
      "grad_norm": 0.0015322680119425058,
      "learning_rate": 5.653175565058888e-07,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 1051
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.573509933774835,
      "grad_norm": 0.006051396485418081,
      "learning_rate": 5.507966677132181e-07,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 1052
    },
    {
      "classification_loss": 0.0001,
      "epoch": 5.578807947019867,
      "grad_norm": 0.008480217307806015,
      "learning_rate": 5.364621030629025e-07,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 1053
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.584105960264901,
      "grad_norm": 0.003863508580252528,
      "learning_rate": 5.223139998806459e-07,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 1054
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.589403973509934,
      "grad_norm": 0.0014740551123395562,
      "learning_rate": 5.083524937058415e-07,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 1055
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.594701986754967,
      "grad_norm": 0.003757091937586665,
      "learning_rate": 4.9457771829027e-07,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 1056
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.6,
      "grad_norm": 0.0010250721825286746,
      "learning_rate": 4.809898055968298e-07,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 1057
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.605298013245033,
      "grad_norm": 0.0014732067938894033,
      "learning_rate": 4.675888857982669e-07,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 1058
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.610596026490066,
      "grad_norm": 0.0013185341376811266,
      "learning_rate": 4.543750872759245e-07,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 1059
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.6158940397351,
      "grad_norm": 0.0038631560746580362,
      "learning_rate": 4.4134853661852215e-07,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 1060
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.621192052980133,
      "grad_norm": 0.0022101786453276873,
      "learning_rate": 4.285093586209299e-07,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 1061
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.626490066225165,
      "grad_norm": 0.0006641183281317353,
      "learning_rate": 4.1585767628298466e-07,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 1062
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.631788079470199,
      "grad_norm": 0.0028428114019334316,
      "learning_rate": 4.033936108083092e-07,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 1063
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.637086092715232,
      "grad_norm": 0.0010977353667840362,
      "learning_rate": 3.911172816031461e-07,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 1064
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.642384105960264,
      "grad_norm": 0.0006411796202883124,
      "learning_rate": 3.7902880627521676e-07,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 1065
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.647682119205298,
      "grad_norm": 0.0026441917289048433,
      "learning_rate": 3.671283006325976e-07,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 1066
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.652980132450331,
      "grad_norm": 0.001259256387129426,
      "learning_rate": 3.554158786826056e-07,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 1067
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.658278145695364,
      "grad_norm": 0.0018623434007167816,
      "learning_rate": 3.4389165263071233e-07,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 1068
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.663576158940398,
      "grad_norm": 0.001959217246621847,
      "learning_rate": 3.32555732879456e-07,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 1069
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.66887417218543,
      "grad_norm": 0.000637912773527205,
      "learning_rate": 3.214082280274067e-07,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 1070
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.674172185430463,
      "grad_norm": 0.0007751178345642984,
      "learning_rate": 3.104492448681007e-07,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 1071
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.679470198675497,
      "grad_norm": 0.0031715852674096823,
      "learning_rate": 2.9967888838903226e-07,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 1072
    },
    {
      "classification_loss": 0.0001,
      "epoch": 5.68476821192053,
      "grad_norm": 0.0059626298025250435,
      "learning_rate": 2.8909726177064776e-07,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 1073
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.6900662251655625,
      "grad_norm": 0.0014221356250345707,
      "learning_rate": 2.787044663853489e-07,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 1074
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.695364238410596,
      "grad_norm": 0.0014999730046838522,
      "learning_rate": 2.68500601796533e-07,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 1075
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.700662251655629,
      "grad_norm": 0.002739256713539362,
      "learning_rate": 2.584857657576234e-07,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 1076
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.705960264900662,
      "grad_norm": 0.0018097941065207124,
      "learning_rate": 2.486600542111539e-07,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 1077
    },
    {
      "classification_loss": 0.0001,
      "epoch": 5.711258278145696,
      "grad_norm": 0.011189226061105728,
      "learning_rate": 2.3902356128783e-07,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 1078
    },
    {
      "classification_loss": 0.0001,
      "epoch": 5.716556291390728,
      "grad_norm": 0.0053149014711380005,
      "learning_rate": 2.2957637930563825e-07,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 1079
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.7218543046357615,
      "grad_norm": 0.002803454641252756,
      "learning_rate": 2.2031859876895602e-07,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 1080
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.727152317880795,
      "grad_norm": 0.002933382522314787,
      "learning_rate": 2.1125030836768977e-07,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 1081
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.732450331125828,
      "grad_norm": 0.0007761417073197663,
      "learning_rate": 2.0237159497642712e-07,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 1082
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.737748344370861,
      "grad_norm": 0.006118531338870525,
      "learning_rate": 1.936825436535905e-07,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 1083
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.743046357615894,
      "grad_norm": 0.002105657709762454,
      "learning_rate": 1.8518323764064038e-07,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 1084
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.748344370860927,
      "grad_norm": 0.0006335449870675802,
      "learning_rate": 1.7687375836126897e-07,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 1085
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.7536423841059605,
      "grad_norm": 0.006882051005959511,
      "learning_rate": 1.6875418542061206e-07,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 1086
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.758940397350994,
      "grad_norm": 0.001206000684760511,
      "learning_rate": 1.6082459660450743e-07,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 1087
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.764238410596026,
      "grad_norm": 0.0033663129433989525,
      "learning_rate": 1.530850678787288e-07,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 1088
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.76953642384106,
      "grad_norm": 0.0021666972897946835,
      "learning_rate": 1.4553567338827068e-07,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 1089
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.774834437086093,
      "grad_norm": 0.0040387557819485664,
      "learning_rate": 1.3817648545663365e-07,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 1090
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.780132450331126,
      "grad_norm": 0.005878349766135216,
      "learning_rate": 1.3100757458512914e-07,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 1091
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.785430463576159,
      "grad_norm": 0.0014043748378753662,
      "learning_rate": 1.2402900945221118e-07,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 1092
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.790728476821192,
      "grad_norm": 0.0029188827611505985,
      "learning_rate": 1.1724085691280806e-07,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 1093
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.796026490066225,
      "grad_norm": 0.009119461290538311,
      "learning_rate": 1.1064318199769387e-07,
      "lm_loss": 0.0,
      "loss": 0.0001,
      "step": 1094
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.801324503311259,
      "grad_norm": 0.0007859122706577182,
      "learning_rate": 1.0423604791285791e-07,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 1095
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.806622516556291,
      "grad_norm": 0.0010608291486278176,
      "learning_rate": 9.801951603889859e-08,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 1096
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.811920529801324,
      "grad_norm": 0.001316172769293189,
      "learning_rate": 9.199364593043936e-08,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 1097
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.817218543046358,
      "grad_norm": 0.0012531770626083016,
      "learning_rate": 8.615849531555365e-08,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 1098
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.822516556291391,
      "grad_norm": 0.0020592291839420795,
      "learning_rate": 8.051412009521864e-08,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 1099
    },
    {
      "classification_loss": 0.0,
      "epoch": 5.827814569536423,
      "grad_norm": 0.0017771337879821658,
      "learning_rate": 7.50605743427757e-08,
      "lm_loss": 0.0,
      "loss": 0.0,
      "step": 1100
    },
    {
      "epoch": 5.827814569536423,
      "eval_accuracy": 0.9211267828941345,
      "eval_auroc": 0.9710983037948608,
      "eval_classification_loss": 0.6522337794303894,
      "eval_f1": 0.9186046719551086,
      "eval_lm_loss": 0.061974067240953445,
      "eval_loss": 0.7142078876495361,
      "eval_precision": 0.9239766001701355,
      "eval_recall": 0.913294792175293,
      "eval_runtime": 59.0104,
      "eval_samples_per_second": 6.016,
      "eval_steps_per_second": 3.016,
      "step": 1100
    }
  ],
  "logging_steps": 1,
  "max_steps": 1128,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 6,
  "save_steps": 50,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 2.3423291801911296e+17,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}