{
  "best_metric": 0.9090909090909091,
  "best_model_checkpoint": "checkpoint-350-4th-march/checkpoint-10",
  "epoch": 89.00555555555556,
  "eval_steps": 500,
  "global_step": 450,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "eval_accuracy": 0.7272727272727273,
      "eval_f1": 0.6727272727272727,
      "eval_loss": 0.6355372667312622,
      "eval_precision": 0.6590909090909091,
      "eval_recall": 0.7272727272727273,
      "eval_runtime": 24.4958,
      "eval_samples_per_second": 0.449,
      "eval_steps_per_second": 0.041,
      "step": 5
    },
    {
      "epoch": 1.01,
      "grad_norm": 5.713123321533203,
      "learning_rate": 2.2222222222222225e-06,
      "loss": 0.2306,
      "step": 10
    },
    {
      "epoch": 1.01,
      "eval_accuracy": 0.9090909090909091,
      "eval_f1": 0.8787878787878788,
      "eval_loss": 0.5734981894493103,
      "eval_precision": 0.8636363636363636,
      "eval_recall": 0.9090909090909091,
      "eval_runtime": 23.8695,
      "eval_samples_per_second": 0.461,
      "eval_steps_per_second": 0.042,
      "step": 10
    },
    {
      "epoch": 2.01,
      "eval_accuracy": 0.8181818181818182,
      "eval_f1": 0.7878787878787878,
      "eval_loss": 0.5388274788856506,
      "eval_precision": 0.7727272727272727,
      "eval_recall": 0.8181818181818182,
      "eval_runtime": 23.9227,
      "eval_samples_per_second": 0.46,
      "eval_steps_per_second": 0.042,
      "step": 15
    },
    {
      "epoch": 3.01,
      "grad_norm": 1.6992627382278442,
      "learning_rate": 4.444444444444445e-06,
      "loss": 0.1604,
      "step": 20
    },
    {
      "epoch": 3.01,
      "eval_accuracy": 0.8181818181818182,
      "eval_f1": 0.7575757575757577,
      "eval_loss": 0.550200879573822,
      "eval_precision": 0.7272727272727273,
      "eval_recall": 0.8181818181818182,
      "eval_runtime": 24.9639,
      "eval_samples_per_second": 0.441,
      "eval_steps_per_second": 0.04,
      "step": 20
    },
    {
      "epoch": 4.01,
      "eval_accuracy": 0.7272727272727273,
      "eval_f1": 0.6727272727272727,
      "eval_loss": 0.6521897912025452,
      "eval_precision": 0.6590909090909091,
      "eval_recall": 0.7272727272727273,
      "eval_runtime": 27.1115,
      "eval_samples_per_second": 0.406,
      "eval_steps_per_second": 0.037,
      "step": 25
    },
    {
      "epoch": 5.01,
      "grad_norm": 4.790561199188232,
      "learning_rate": 6.666666666666667e-06,
      "loss": 0.1979,
      "step": 30
    },
    {
      "epoch": 5.01,
      "eval_accuracy": 0.7272727272727273,
      "eval_f1": 0.6727272727272727,
      "eval_loss": 0.8280015587806702,
      "eval_precision": 0.6590909090909091,
      "eval_recall": 0.7272727272727273,
      "eval_runtime": 28.3405,
      "eval_samples_per_second": 0.388,
      "eval_steps_per_second": 0.035,
      "step": 30
    },
    {
      "epoch": 6.01,
      "eval_accuracy": 0.7272727272727273,
      "eval_f1": 0.6727272727272727,
      "eval_loss": 0.6641684174537659,
      "eval_precision": 0.6590909090909091,
      "eval_recall": 0.7272727272727273,
      "eval_runtime": 27.3051,
      "eval_samples_per_second": 0.403,
      "eval_steps_per_second": 0.037,
      "step": 35
    },
    {
      "epoch": 7.01,
      "grad_norm": 8.951165199279785,
      "learning_rate": 8.88888888888889e-06,
      "loss": 0.1854,
      "step": 40
    },
    {
      "epoch": 7.01,
      "eval_accuracy": 0.8181818181818182,
      "eval_f1": 0.7575757575757577,
      "eval_loss": 0.5681249499320984,
      "eval_precision": 0.7272727272727273,
      "eval_recall": 0.8181818181818182,
      "eval_runtime": 26.9304,
      "eval_samples_per_second": 0.408,
      "eval_steps_per_second": 0.037,
      "step": 40
    },
    {
      "epoch": 8.01,
      "eval_accuracy": 0.8181818181818182,
      "eval_f1": 0.7727272727272727,
      "eval_loss": 0.5571709275245667,
      "eval_precision": 0.7575757575757577,
      "eval_recall": 0.8181818181818182,
      "eval_runtime": 27.0313,
      "eval_samples_per_second": 0.407,
      "eval_steps_per_second": 0.037,
      "step": 45
    },
    {
      "epoch": 9.01,
      "grad_norm": 11.677026748657227,
      "learning_rate": 1.1111111111111112e-05,
      "loss": 0.1513,
      "step": 50
    },
    {
      "epoch": 9.01,
      "eval_accuracy": 0.7272727272727273,
      "eval_f1": 0.6727272727272727,
      "eval_loss": 0.6411817669868469,
      "eval_precision": 0.6590909090909091,
      "eval_recall": 0.7272727272727273,
      "eval_runtime": 21.8481,
      "eval_samples_per_second": 0.503,
      "eval_steps_per_second": 0.046,
      "step": 50
    },
    {
      "epoch": 10.01,
      "eval_accuracy": 0.7272727272727273,
      "eval_f1": 0.6727272727272727,
      "eval_loss": 0.6945109963417053,
      "eval_precision": 0.6590909090909091,
      "eval_recall": 0.7272727272727273,
      "eval_runtime": 20.9102,
      "eval_samples_per_second": 0.526,
      "eval_steps_per_second": 0.048,
      "step": 55
    },
    {
      "epoch": 11.01,
      "grad_norm": 4.137542724609375,
      "learning_rate": 1.3333333333333333e-05,
      "loss": 0.1214,
      "step": 60
    },
    {
      "epoch": 11.01,
      "eval_accuracy": 0.7272727272727273,
      "eval_f1": 0.6515151515151515,
      "eval_loss": 0.7864866852760315,
      "eval_precision": 0.6212121212121212,
      "eval_recall": 0.7272727272727273,
      "eval_runtime": 20.5893,
      "eval_samples_per_second": 0.534,
      "eval_steps_per_second": 0.049,
      "step": 60
    },
    {
      "epoch": 12.01,
      "eval_accuracy": 0.7272727272727273,
      "eval_f1": 0.6727272727272727,
      "eval_loss": 0.7256456017494202,
      "eval_precision": 0.6590909090909091,
      "eval_recall": 0.7272727272727273,
      "eval_runtime": 21.3156,
      "eval_samples_per_second": 0.516,
      "eval_steps_per_second": 0.047,
      "step": 65
    },
    {
      "epoch": 13.01,
      "grad_norm": 3.5624358654022217,
      "learning_rate": 1.5555555555555555e-05,
      "loss": 0.1661,
      "step": 70
    },
    {
      "epoch": 13.01,
      "eval_accuracy": 0.8181818181818182,
      "eval_f1": 0.7575757575757577,
      "eval_loss": 0.5460242629051208,
      "eval_precision": 0.7272727272727273,
      "eval_recall": 0.8181818181818182,
      "eval_runtime": 21.5745,
      "eval_samples_per_second": 0.51,
      "eval_steps_per_second": 0.046,
      "step": 70
    },
    {
      "epoch": 14.01,
      "eval_accuracy": 0.8181818181818182,
      "eval_f1": 0.7575757575757577,
      "eval_loss": 0.5317673087120056,
      "eval_precision": 0.7272727272727273,
      "eval_recall": 0.8181818181818182,
      "eval_runtime": 21.1185,
      "eval_samples_per_second": 0.521,
      "eval_steps_per_second": 0.047,
      "step": 75
    },
    {
      "epoch": 15.01,
      "grad_norm": 1.1072447299957275,
      "learning_rate": 1.777777777777778e-05,
      "loss": 0.1847,
      "step": 80
    },
    {
      "epoch": 15.01,
      "eval_accuracy": 0.8181818181818182,
      "eval_f1": 0.7727272727272727,
      "eval_loss": 0.5134172439575195,
      "eval_precision": 0.7575757575757577,
      "eval_recall": 0.8181818181818182,
      "eval_runtime": 20.7396,
      "eval_samples_per_second": 0.53,
      "eval_steps_per_second": 0.048,
      "step": 80
    },
    {
      "epoch": 16.01,
      "eval_accuracy": 0.9090909090909091,
      "eval_f1": 0.8787878787878788,
      "eval_loss": 0.4267767071723938,
      "eval_precision": 0.8636363636363636,
      "eval_recall": 0.9090909090909091,
      "eval_runtime": 20.5432,
      "eval_samples_per_second": 0.535,
      "eval_steps_per_second": 0.049,
      "step": 85
    },
    {
      "epoch": 17.01,
      "grad_norm": 13.903785705566406,
      "learning_rate": 2e-05,
      "loss": 0.3083,
      "step": 90
    },
    {
      "epoch": 17.01,
      "eval_accuracy": 0.7272727272727273,
      "eval_f1": 0.6666666666666666,
      "eval_loss": 0.7989960312843323,
      "eval_precision": 0.6363636363636364,
      "eval_recall": 0.7272727272727273,
      "eval_runtime": 20.5742,
      "eval_samples_per_second": 0.535,
      "eval_steps_per_second": 0.049,
      "step": 90
    },
    {
      "epoch": 18.01,
      "eval_accuracy": 0.7272727272727273,
      "eval_f1": 0.6666666666666666,
      "eval_loss": 0.8174381256103516,
      "eval_precision": 0.6363636363636364,
      "eval_recall": 0.7272727272727273,
      "eval_runtime": 20.4736,
      "eval_samples_per_second": 0.537,
      "eval_steps_per_second": 0.049,
      "step": 95
    },
    {
      "epoch": 19.01,
      "grad_norm": 0.5411530137062073,
      "learning_rate": 2.2222222222222223e-05,
      "loss": 0.2193,
      "step": 100
    },
    {
      "epoch": 19.01,
      "eval_accuracy": 0.7272727272727273,
      "eval_f1": 0.6727272727272727,
      "eval_loss": 0.5629628896713257,
      "eval_precision": 0.6590909090909091,
      "eval_recall": 0.7272727272727273,
      "eval_runtime": 23.091,
      "eval_samples_per_second": 0.476,
      "eval_steps_per_second": 0.043,
      "step": 100
    },
    {
      "epoch": 20.01,
      "eval_accuracy": 0.8181818181818182,
      "eval_f1": 0.7575757575757577,
      "eval_loss": 0.452359676361084,
      "eval_precision": 0.7272727272727273,
      "eval_recall": 0.8181818181818182,
      "eval_runtime": 22.8931,
      "eval_samples_per_second": 0.48,
      "eval_steps_per_second": 0.044,
      "step": 105
    },
    {
      "epoch": 21.01,
      "grad_norm": 6.751333713531494,
      "learning_rate": 2.4444444444444445e-05,
      "loss": 0.1367,
      "step": 110
    },
    {
      "epoch": 21.01,
      "eval_accuracy": 0.8181818181818182,
      "eval_f1": 0.7575757575757577,
      "eval_loss": 0.588365375995636,
      "eval_precision": 0.7272727272727273,
      "eval_recall": 0.8181818181818182,
      "eval_runtime": 22.2164,
      "eval_samples_per_second": 0.495,
      "eval_steps_per_second": 0.045,
      "step": 110
    },
    {
      "epoch": 22.01,
      "eval_accuracy": 0.7272727272727273,
      "eval_f1": 0.6818181818181818,
      "eval_loss": 0.6520605683326721,
      "eval_precision": 0.6666666666666666,
      "eval_recall": 0.7272727272727273,
      "eval_runtime": 21.6145,
      "eval_samples_per_second": 0.509,
      "eval_steps_per_second": 0.046,
      "step": 115
    },
    {
      "epoch": 23.01,
      "grad_norm": 2.2240047454833984,
      "learning_rate": 2.6666666666666667e-05,
      "loss": 0.1478,
      "step": 120
    },
    {
      "epoch": 23.01,
      "eval_accuracy": 0.8181818181818182,
      "eval_f1": 0.7575757575757577,
      "eval_loss": 0.550651490688324,
      "eval_precision": 0.7272727272727273,
      "eval_recall": 0.8181818181818182,
      "eval_runtime": 22.2994,
      "eval_samples_per_second": 0.493,
      "eval_steps_per_second": 0.045,
      "step": 120
    },
    {
      "epoch": 24.01,
      "eval_accuracy": 0.8181818181818182,
      "eval_f1": 0.7727272727272727,
      "eval_loss": 0.5914822220802307,
      "eval_precision": 0.7575757575757577,
      "eval_recall": 0.8181818181818182,
      "eval_runtime": 22.4716,
      "eval_samples_per_second": 0.49,
      "eval_steps_per_second": 0.045,
      "step": 125
    },
    {
      "epoch": 25.01,
      "grad_norm": 2.0005195140838623,
      "learning_rate": 2.8888888888888888e-05,
      "loss": 0.173,
      "step": 130
    },
    {
      "epoch": 25.01,
      "eval_accuracy": 0.9090909090909091,
      "eval_f1": 0.8787878787878788,
      "eval_loss": 0.36792269349098206,
      "eval_precision": 0.8636363636363636,
      "eval_recall": 0.9090909090909091,
      "eval_runtime": 22.1546,
      "eval_samples_per_second": 0.497,
      "eval_steps_per_second": 0.045,
      "step": 130
    },
    {
      "epoch": 26.01,
      "eval_accuracy": 0.8181818181818182,
      "eval_f1": 0.7878787878787878,
      "eval_loss": 0.3868860900402069,
      "eval_precision": 0.7727272727272727,
      "eval_recall": 0.8181818181818182,
      "eval_runtime": 21.2578,
      "eval_samples_per_second": 0.517,
      "eval_steps_per_second": 0.047,
      "step": 135
    },
    {
      "epoch": 27.01,
      "grad_norm": 7.4427385330200195,
      "learning_rate": 3.111111111111111e-05,
      "loss": 0.2515,
      "step": 140
    },
    {
      "epoch": 27.01,
      "eval_accuracy": 0.7272727272727273,
      "eval_f1": 0.696969696969697,
      "eval_loss": 0.4347504675388336,
      "eval_precision": 0.6818181818181818,
      "eval_recall": 0.7272727272727273,
      "eval_runtime": 23.2343,
      "eval_samples_per_second": 0.473,
      "eval_steps_per_second": 0.043,
      "step": 140
    },
    {
      "epoch": 28.01,
      "eval_accuracy": 0.8181818181818182,
      "eval_f1": 0.7727272727272727,
      "eval_loss": 0.5722665786743164,
      "eval_precision": 0.7575757575757577,
      "eval_recall": 0.8181818181818182,
      "eval_runtime": 20.9817,
      "eval_samples_per_second": 0.524,
      "eval_steps_per_second": 0.048,
      "step": 145
    },
    {
      "epoch": 29.01,
      "grad_norm": 3.472019672393799,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 0.1947,
      "step": 150
    },
    {
      "epoch": 29.01,
      "eval_accuracy": 0.7272727272727273,
      "eval_f1": 0.6727272727272727,
      "eval_loss": 0.824672520160675,
      "eval_precision": 0.6590909090909091,
      "eval_recall": 0.7272727272727273,
      "eval_runtime": 20.2976,
      "eval_samples_per_second": 0.542,
      "eval_steps_per_second": 0.049,
      "step": 150
    },
    {
      "epoch": 30.01,
      "eval_accuracy": 0.8181818181818182,
      "eval_f1": 0.7575757575757577,
      "eval_loss": 0.4699678421020508,
      "eval_precision": 0.7272727272727273,
      "eval_recall": 0.8181818181818182,
      "eval_runtime": 19.7939,
      "eval_samples_per_second": 0.556,
      "eval_steps_per_second": 0.051,
      "step": 155
    },
    {
      "epoch": 31.01,
      "grad_norm": 8.735132217407227,
      "learning_rate": 3.555555555555556e-05,
      "loss": 0.1871,
      "step": 160
    },
    {
      "epoch": 31.01,
      "eval_accuracy": 0.7272727272727273,
      "eval_f1": 0.6515151515151515,
      "eval_loss": 0.6745826601982117,
      "eval_precision": 0.6212121212121212,
      "eval_recall": 0.7272727272727273,
      "eval_runtime": 21.2096,
      "eval_samples_per_second": 0.519,
      "eval_steps_per_second": 0.047,
      "step": 160
    },
    {
      "epoch": 32.01,
      "eval_accuracy": 0.7272727272727273,
      "eval_f1": 0.6454545454545454,
      "eval_loss": 1.3604460954666138,
      "eval_precision": 0.606060606060606,
      "eval_recall": 0.7272727272727273,
      "eval_runtime": 20.7259,
      "eval_samples_per_second": 0.531,
      "eval_steps_per_second": 0.048,
      "step": 165
    },
    {
      "epoch": 33.01,
      "grad_norm": 1.2702025175094604,
      "learning_rate": 3.777777777777778e-05,
      "loss": 0.1651,
      "step": 170
    },
    {
      "epoch": 33.01,
      "eval_accuracy": 0.7272727272727273,
      "eval_f1": 0.6666666666666666,
      "eval_loss": 0.786778450012207,
      "eval_precision": 0.6363636363636364,
      "eval_recall": 0.7272727272727273,
      "eval_runtime": 22.3289,
      "eval_samples_per_second": 0.493,
      "eval_steps_per_second": 0.045,
      "step": 170
    },
    {
      "epoch": 34.01,
      "eval_accuracy": 0.8181818181818182,
      "eval_f1": 0.7575757575757577,
      "eval_loss": 0.7492481470108032,
      "eval_precision": 0.7272727272727273,
      "eval_recall": 0.8181818181818182,
      "eval_runtime": 20.2554,
      "eval_samples_per_second": 0.543,
      "eval_steps_per_second": 0.049,
      "step": 175
    },
    {
      "epoch": 35.01,
      "grad_norm": 1.1660475730895996,
      "learning_rate": 4e-05,
      "loss": 0.1621,
      "step": 180
    },
    {
      "epoch": 35.01,
      "eval_accuracy": 0.7272727272727273,
      "eval_f1": 0.6727272727272727,
      "eval_loss": 0.707514762878418,
      "eval_precision": 0.6590909090909091,
      "eval_recall": 0.7272727272727273,
      "eval_runtime": 20.9293,
      "eval_samples_per_second": 0.526,
      "eval_steps_per_second": 0.048,
      "step": 180
    },
    {
      "epoch": 36.01,
      "eval_accuracy": 0.8181818181818182,
      "eval_f1": 0.7727272727272727,
      "eval_loss": 0.5149413347244263,
      "eval_precision": 0.7575757575757577,
      "eval_recall": 0.8181818181818182,
      "eval_runtime": 20.6993,
      "eval_samples_per_second": 0.531,
      "eval_steps_per_second": 0.048,
      "step": 185
    },
    {
      "epoch": 37.01,
      "grad_norm": 2.7489700317382812,
      "learning_rate": 4.222222222222222e-05,
      "loss": 0.094,
      "step": 190
    },
    {
      "epoch": 37.01,
      "eval_accuracy": 0.8181818181818182,
      "eval_f1": 0.7575757575757577,
      "eval_loss": 0.6731873750686646,
      "eval_precision": 0.7272727272727273,
      "eval_recall": 0.8181818181818182,
      "eval_runtime": 19.5579,
      "eval_samples_per_second": 0.562,
      "eval_steps_per_second": 0.051,
      "step": 190
    },
    {
      "epoch": 38.01,
      "eval_accuracy": 0.8181818181818182,
      "eval_f1": 0.7575757575757577,
      "eval_loss": 0.7046161890029907,
      "eval_precision": 0.7272727272727273,
      "eval_recall": 0.8181818181818182,
      "eval_runtime": 20.2638,
      "eval_samples_per_second": 0.543,
      "eval_steps_per_second": 0.049,
      "step": 195
    },
    {
      "epoch": 39.01,
      "grad_norm": 3.9047691822052,
      "learning_rate": 4.4444444444444447e-05,
      "loss": 0.1494,
      "step": 200
    },
    {
      "epoch": 39.01,
      "eval_accuracy": 0.8181818181818182,
      "eval_f1": 0.7727272727272727,
      "eval_loss": 0.667763888835907,
      "eval_precision": 0.7575757575757577,
      "eval_recall": 0.8181818181818182,
      "eval_runtime": 21.6759,
      "eval_samples_per_second": 0.507,
      "eval_steps_per_second": 0.046,
      "step": 200
    },
    {
      "epoch": 40.01,
      "eval_accuracy": 0.8181818181818182,
      "eval_f1": 0.8181818181818182,
      "eval_loss": 0.4195019602775574,
      "eval_precision": 0.8181818181818182,
      "eval_recall": 0.8181818181818182,
      "eval_runtime": 21.8447,
      "eval_samples_per_second": 0.504,
      "eval_steps_per_second": 0.046,
      "step": 205
    },
    {
      "epoch": 41.01,
      "grad_norm": 20.463916778564453,
      "learning_rate": 4.666666666666667e-05,
      "loss": 0.1819,
      "step": 210
    },
    {
      "epoch": 41.01,
      "eval_accuracy": 0.8181818181818182,
      "eval_f1": 0.7575757575757577,
      "eval_loss": 0.48133182525634766,
      "eval_precision": 0.7272727272727273,
      "eval_recall": 0.8181818181818182,
      "eval_runtime": 20.7381,
      "eval_samples_per_second": 0.53,
      "eval_steps_per_second": 0.048,
      "step": 210
    },
    {
      "epoch": 42.01,
      "eval_accuracy": 0.6363636363636364,
      "eval_f1": 0.5757575757575758,
      "eval_loss": 1.2362233400344849,
      "eval_precision": 0.5984848484848485,
      "eval_recall": 0.6363636363636364,
      "eval_runtime": 20.9607,
      "eval_samples_per_second": 0.525,
      "eval_steps_per_second": 0.048,
      "step": 215
    },
    {
      "epoch": 43.01,
      "grad_norm": 20.663036346435547,
      "learning_rate": 4.888888888888889e-05,
      "loss": 0.3123,
      "step": 220
    },
    {
      "epoch": 43.01,
      "eval_accuracy": 0.6363636363636364,
      "eval_f1": 0.5848484848484848,
      "eval_loss": 1.2863417863845825,
      "eval_precision": 0.606060606060606,
      "eval_recall": 0.6363636363636364,
      "eval_runtime": 24.9545,
      "eval_samples_per_second": 0.441,
      "eval_steps_per_second": 0.04,
      "step": 220
    },
    {
      "epoch": 44.01,
      "eval_accuracy": 0.6363636363636364,
      "eval_f1": 0.5696969696969697,
      "eval_loss": 1.4153846502304077,
      "eval_precision": 0.5757575757575757,
      "eval_recall": 0.6363636363636364,
      "eval_runtime": 21.2069,
      "eval_samples_per_second": 0.519,
      "eval_steps_per_second": 0.047,
      "step": 225
    },
    {
      "epoch": 45.01,
      "grad_norm": 11.3672456741333,
      "learning_rate": 4.962962962962963e-05,
      "loss": 0.3182,
      "step": 230
    },
    {
      "epoch": 45.01,
      "eval_accuracy": 0.7272727272727273,
      "eval_f1": 0.6818181818181818,
      "eval_loss": 0.6307955384254456,
      "eval_precision": 0.6666666666666666,
      "eval_recall": 0.7272727272727273,
      "eval_runtime": 20.1678,
      "eval_samples_per_second": 0.545,
      "eval_steps_per_second": 0.05,
      "step": 230
    },
    {
      "epoch": 46.01,
      "eval_accuracy": 0.8181818181818182,
      "eval_f1": 0.7575757575757577,
      "eval_loss": 0.4325122535228729,
      "eval_precision": 0.7272727272727273,
      "eval_recall": 0.8181818181818182,
      "eval_runtime": 21.2567,
      "eval_samples_per_second": 0.517,
      "eval_steps_per_second": 0.047,
      "step": 235
    },
    {
      "epoch": 47.01,
      "grad_norm": 17.50649642944336,
      "learning_rate": 4.888888888888889e-05,
      "loss": 0.1992,
      "step": 240
    },
    {
      "epoch": 47.01,
      "eval_accuracy": 0.9090909090909091,
      "eval_f1": 0.8787878787878788,
      "eval_loss": 0.42543408274650574,
      "eval_precision": 0.8636363636363636,
      "eval_recall": 0.9090909090909091,
      "eval_runtime": 20.0391,
      "eval_samples_per_second": 0.549,
      "eval_steps_per_second": 0.05,
      "step": 240
    },
    {
      "epoch": 48.01,
      "eval_accuracy": 0.8181818181818182,
      "eval_f1": 0.7818181818181817,
      "eval_loss": 0.5966838598251343,
      "eval_precision": 0.8030303030303029,
      "eval_recall": 0.8181818181818182,
      "eval_runtime": 20.4865,
      "eval_samples_per_second": 0.537,
      "eval_steps_per_second": 0.049,
      "step": 245
    },
    {
      "epoch": 49.01,
      "grad_norm": 10.43728256225586,
      "learning_rate": 4.814814814814815e-05,
      "loss": 0.1887,
      "step": 250
    },
    {
      "epoch": 49.01,
      "eval_accuracy": 0.9090909090909091,
      "eval_f1": 0.8787878787878788,
      "eval_loss": 0.4350906312465668,
      "eval_precision": 0.8636363636363636,
      "eval_recall": 0.9090909090909091,
      "eval_runtime": 20.3501,
      "eval_samples_per_second": 0.541,
      "eval_steps_per_second": 0.049,
      "step": 250
    },
    {
      "epoch": 50.01,
      "eval_accuracy": 0.8181818181818182,
      "eval_f1": 0.8181818181818182,
      "eval_loss": 0.5225472450256348,
      "eval_precision": 0.8181818181818182,
      "eval_recall": 0.8181818181818182,
      "eval_runtime": 20.4273,
      "eval_samples_per_second": 0.538,
      "eval_steps_per_second": 0.049,
      "step": 255
    },
    {
      "epoch": 51.01,
      "grad_norm": 11.036069869995117,
      "learning_rate": 4.740740740740741e-05,
      "loss": 0.2046,
      "step": 260
    },
    {
      "epoch": 51.01,
      "eval_accuracy": 0.7272727272727273,
      "eval_f1": 0.6818181818181818,
      "eval_loss": 0.6759114861488342,
      "eval_precision": 0.6666666666666667,
      "eval_recall": 0.7272727272727273,
      "eval_runtime": 20.7436,
      "eval_samples_per_second": 0.53,
      "eval_steps_per_second": 0.048,
      "step": 260
    },
    {
      "epoch": 52.01,
      "eval_accuracy": 0.8181818181818182,
      "eval_f1": 0.7575757575757577,
      "eval_loss": 0.5574895739555359,
      "eval_precision": 0.7272727272727273,
      "eval_recall": 0.8181818181818182,
      "eval_runtime": 20.5552,
      "eval_samples_per_second": 0.535,
      "eval_steps_per_second": 0.049,
      "step": 265
    },
    {
      "epoch": 53.01,
      "grad_norm": 9.632101058959961,
      "learning_rate": 4.666666666666667e-05,
      "loss": 0.2225,
      "step": 270
    },
    {
      "epoch": 53.01,
      "eval_accuracy": 0.6363636363636364,
      "eval_f1": 0.5281385281385281,
      "eval_loss": 1.1722840070724487,
      "eval_precision": 0.4818181818181818,
      "eval_recall": 0.6363636363636364,
      "eval_runtime": 25.8215,
      "eval_samples_per_second": 0.426,
      "eval_steps_per_second": 0.039,
      "step": 270
    },
    {
      "epoch": 54.01,
      "eval_accuracy": 0.7272727272727273,
      "eval_f1": 0.6818181818181818,
      "eval_loss": 0.9067409634590149,
      "eval_precision": 0.6666666666666666,
      "eval_recall": 0.7272727272727273,
      "eval_runtime": 21.6949,
      "eval_samples_per_second": 0.507,
      "eval_steps_per_second": 0.046,
      "step": 275
    },
    {
      "epoch": 55.01,
      "grad_norm": 18.024147033691406,
      "learning_rate": 4.592592592592593e-05,
      "loss": 0.2867,
      "step": 280
    },
    {
      "epoch": 55.01,
      "eval_accuracy": 0.9090909090909091,
      "eval_f1": 0.8787878787878788,
      "eval_loss": 0.29047486186027527,
      "eval_precision": 0.8636363636363636,
      "eval_recall": 0.9090909090909091,
      "eval_runtime": 20.839,
      "eval_samples_per_second": 0.528,
      "eval_steps_per_second": 0.048,
      "step": 280
    },
    {
      "epoch": 56.01,
      "eval_accuracy": 0.7272727272727273,
      "eval_f1": 0.696969696969697,
      "eval_loss": 0.6336184144020081,
      "eval_precision": 0.6818181818181818,
      "eval_recall": 0.7272727272727273,
      "eval_runtime": 22.0174,
      "eval_samples_per_second": 0.5,
      "eval_steps_per_second": 0.045,
      "step": 285
    },
    {
      "epoch": 57.01,
      "grad_norm": 1.8721508979797363,
      "learning_rate": 4.518518518518519e-05,
      "loss": 0.2137,
      "step": 290
    },
    {
      "epoch": 57.01,
      "eval_accuracy": 0.8181818181818182,
      "eval_f1": 0.7575757575757577,
      "eval_loss": 0.5004566311836243,
      "eval_precision": 0.7272727272727273,
      "eval_recall": 0.8181818181818182,
      "eval_runtime": 22.896,
      "eval_samples_per_second": 0.48,
      "eval_steps_per_second": 0.044,
      "step": 290
    },
    {
      "epoch": 58.01,
      "eval_accuracy": 0.8181818181818182,
      "eval_f1": 0.7575757575757577,
      "eval_loss": 0.4470617175102234,
      "eval_precision": 0.7272727272727273,
      "eval_recall": 0.8181818181818182,
      "eval_runtime": 20.6595,
      "eval_samples_per_second": 0.532,
      "eval_steps_per_second": 0.048,
      "step": 295
    },
    {
      "epoch": 59.01,
      "grad_norm": 1.4810432195663452,
      "learning_rate": 4.4444444444444447e-05,
      "loss": 0.2609,
      "step": 300
    },
    {
      "epoch": 59.01,
      "eval_accuracy": 0.7272727272727273,
      "eval_f1": 0.696969696969697,
      "eval_loss": 0.5465022921562195,
      "eval_precision": 0.6818181818181818,
      "eval_recall": 0.7272727272727273,
      "eval_runtime": 20.8964,
      "eval_samples_per_second": 0.526,
      "eval_steps_per_second": 0.048,
      "step": 300
    },
    {
      "epoch": 60.01,
      "eval_accuracy": 0.7272727272727273,
      "eval_f1": 0.696969696969697,
      "eval_loss": 0.5460771918296814,
      "eval_precision": 0.6818181818181818,
      "eval_recall": 0.7272727272727273,
      "eval_runtime": 19.8782,
      "eval_samples_per_second": 0.553,
      "eval_steps_per_second": 0.05,
      "step": 305
    },
    {
      "epoch": 61.01,
      "grad_norm": 10.204845428466797,
      "learning_rate": 4.3703703703703705e-05,
      "loss": 0.1033,
      "step": 310
    },
    {
      "epoch": 61.01,
      "eval_accuracy": 0.8181818181818182,
      "eval_f1": 0.7727272727272727,
      "eval_loss": 0.46728765964508057,
      "eval_precision": 0.7575757575757577,
      "eval_recall": 0.8181818181818182,
      "eval_runtime": 20.0013,
      "eval_samples_per_second": 0.55,
      "eval_steps_per_second": 0.05,
      "step": 310
    },
    {
      "epoch": 62.01,
      "eval_accuracy": 0.8181818181818182,
      "eval_f1": 0.7727272727272727,
      "eval_loss": 0.6281985640525818,
      "eval_precision": 0.7575757575757577,
      "eval_recall": 0.8181818181818182,
      "eval_runtime": 20.7864,
      "eval_samples_per_second": 0.529,
      "eval_steps_per_second": 0.048,
      "step": 315
    },
    {
      "epoch": 63.01,
      "grad_norm": 3.3419110774993896,
      "learning_rate": 4.296296296296296e-05,
      "loss": 0.2027,
      "step": 320
    },
    {
      "epoch": 63.01,
      "eval_accuracy": 0.6363636363636364,
      "eval_f1": 0.5848484848484848,
      "eval_loss": 1.3454620838165283,
      "eval_precision": 0.606060606060606,
      "eval_recall": 0.6363636363636364,
      "eval_runtime": 20.7754,
      "eval_samples_per_second": 0.529,
      "eval_steps_per_second": 0.048,
      "step": 320
    },
    {
      "epoch": 64.01,
      "eval_accuracy": 0.6363636363636364,
      "eval_f1": 0.5848484848484848,
      "eval_loss": 1.2102134227752686,
      "eval_precision": 0.606060606060606,
      "eval_recall": 0.6363636363636364,
      "eval_runtime": 20.3871,
      "eval_samples_per_second": 0.54,
      "eval_steps_per_second": 0.049,
      "step": 325
    },
    {
      "epoch": 65.01,
      "grad_norm": 6.031401634216309,
      "learning_rate": 4.222222222222222e-05,
      "loss": 0.2333,
      "step": 330
    },
    {
      "epoch": 65.01,
      "eval_accuracy": 0.7272727272727273,
      "eval_f1": 0.6606060606060606,
      "eval_loss": 0.8988341689109802,
      "eval_precision": 0.6666666666666666,
      "eval_recall": 0.7272727272727273,
      "eval_runtime": 19.8751,
      "eval_samples_per_second": 0.553,
      "eval_steps_per_second": 0.05,
      "step": 330
    },
    {
      "epoch": 66.01,
      "eval_accuracy": 0.6363636363636364,
      "eval_f1": 0.5757575757575758,
      "eval_loss": 1.21355140209198,
      "eval_precision": 0.5984848484848485,
      "eval_recall": 0.6363636363636364,
      "eval_runtime": 21.2698,
      "eval_samples_per_second": 0.517,
      "eval_steps_per_second": 0.047,
      "step": 335
    },
    {
      "epoch": 67.01,
      "grad_norm": 1.2481231689453125,
      "learning_rate": 4.148148148148148e-05,
      "loss": 0.2014,
      "step": 340
    },
    {
      "epoch": 67.01,
      "eval_accuracy": 0.8181818181818182,
      "eval_f1": 0.7727272727272727,
      "eval_loss": 0.7183884978294373,
      "eval_precision": 0.7575757575757577,
      "eval_recall": 0.8181818181818182,
      "eval_runtime": 20.3008,
      "eval_samples_per_second": 0.542,
      "eval_steps_per_second": 0.049,
      "step": 340
    },
    {
      "epoch": 68.01,
      "eval_accuracy": 0.8181818181818182,
      "eval_f1": 0.7878787878787878,
      "eval_loss": 0.5583562254905701,
      "eval_precision": 0.7727272727272727,
      "eval_recall": 0.8181818181818182,
      "eval_runtime": 22.3248,
      "eval_samples_per_second": 0.493,
      "eval_steps_per_second": 0.045,
      "step": 345
    },
    {
      "epoch": 69.01,
      "grad_norm": 1.0424888134002686,
      "learning_rate": 4.074074074074074e-05,
      "loss": 0.1532,
      "step": 350
    },
    {
      "epoch": 69.01,
      "eval_accuracy": 0.7272727272727273,
      "eval_f1": 0.6666666666666666,
      "eval_loss": 0.9030601382255554,
      "eval_precision": 0.6363636363636364,
      "eval_recall": 0.7272727272727273,
      "eval_runtime": 21.9747,
      "eval_samples_per_second": 0.501,
      "eval_steps_per_second": 0.046,
      "step": 350
    },
    {
      "epoch": 70.01,
      "eval_accuracy": 0.7272727272727273,
      "eval_f1": 0.6515151515151515,
      "eval_loss": 1.1362484693527222,
      "eval_precision": 0.6212121212121212,
      "eval_recall": 0.7272727272727273,
      "eval_runtime": 21.0027,
      "eval_samples_per_second": 0.524,
      "eval_steps_per_second": 0.048,
      "step": 355
    },
    {
      "epoch": 71.01,
      "grad_norm": 13.485044479370117,
      "learning_rate": 4e-05,
      "loss": 0.1102,
      "step": 360
    },
    {
      "epoch": 71.01,
      "eval_accuracy": 0.7272727272727273,
      "eval_f1": 0.6969696969696969,
      "eval_loss": 0.8755394220352173,
      "eval_precision": 0.6818181818181818,
      "eval_recall": 0.7272727272727273,
      "eval_runtime": 20.8601,
      "eval_samples_per_second": 0.527,
      "eval_steps_per_second": 0.048,
      "step": 360
    },
    {
      "epoch": 72.01,
      "eval_accuracy": 0.8181818181818182,
      "eval_f1": 0.7878787878787878,
      "eval_loss": 0.7721637487411499,
      "eval_precision": 0.7727272727272727,
      "eval_recall": 0.8181818181818182,
      "eval_runtime": 20.6398,
      "eval_samples_per_second": 0.533,
      "eval_steps_per_second": 0.048,
      "step": 365
    },
    {
      "epoch": 73.01,
      "grad_norm": 20.92743492126465,
      "learning_rate": 3.925925925925926e-05,
      "loss": 0.2666,
      "step": 370
    },
    {
      "epoch": 73.01,
      "eval_accuracy": 0.8181818181818182,
      "eval_f1": 0.7878787878787878,
      "eval_loss": 0.9053316712379456,
      "eval_precision": 0.7727272727272727,
      "eval_recall": 0.8181818181818182,
      "eval_runtime": 20.7791,
      "eval_samples_per_second": 0.529,
      "eval_steps_per_second": 0.048,
      "step": 370
    },
    {
      "epoch": 74.01,
      "eval_accuracy": 0.8181818181818182,
      "eval_f1": 0.7878787878787878,
      "eval_loss": 0.6602973341941833,
      "eval_precision": 0.7727272727272727,
      "eval_recall": 0.8181818181818182,
      "eval_runtime": 21.2632,
      "eval_samples_per_second": 0.517,
      "eval_steps_per_second": 0.047,
      "step": 375
    },
    {
      "epoch": 75.01,
      "grad_norm": 5.458927631378174,
      "learning_rate": 3.851851851851852e-05,
      "loss": 0.2473,
      "step": 380
    },
    {
      "epoch": 75.01,
      "eval_accuracy": 0.7272727272727273,
      "eval_f1": 0.6818181818181818,
      "eval_loss": 0.5296440720558167,
      "eval_precision": 0.6666666666666666,
      "eval_recall": 0.7272727272727273,
      "eval_runtime": 21.1347,
      "eval_samples_per_second": 0.52,
      "eval_steps_per_second": 0.047,
      "step": 380
    },
    {
      "epoch": 76.01,
      "eval_accuracy": 0.7272727272727273,
      "eval_f1": 0.6818181818181818,
      "eval_loss": 0.5802969336509705,
      "eval_precision": 0.6666666666666667,
      "eval_recall": 0.7272727272727273,
      "eval_runtime": 21.0086,
      "eval_samples_per_second": 0.524,
      "eval_steps_per_second": 0.048,
      "step": 385
    },
    {
      "epoch": 77.01,
      "grad_norm": 6.091598033905029,
      "learning_rate": 3.777777777777778e-05,
      "loss": 0.2011,
      "step": 390
    },
    {
      "epoch": 77.01,
      "eval_accuracy": 0.7272727272727273,
      "eval_f1": 0.6606060606060605,
      "eval_loss": 0.83270663022995,
      "eval_precision": 0.6666666666666666,
      "eval_recall": 0.7272727272727273,
      "eval_runtime": 21.0559,
      "eval_samples_per_second": 0.522,
      "eval_steps_per_second": 0.047,
      "step": 390
    },
    {
      "epoch": 78.01,
      "eval_accuracy": 0.8181818181818182,
      "eval_f1": 0.7575757575757577,
      "eval_loss": 0.4885137677192688,
      "eval_precision": 0.7272727272727273,
      "eval_recall": 0.8181818181818182,
      "eval_runtime": 20.5482,
      "eval_samples_per_second": 0.535,
      "eval_steps_per_second": 0.049,
      "step": 395
    },
    {
      "epoch": 79.01,
      "grad_norm": 67.30104064941406,
      "learning_rate": 3.7037037037037037e-05,
      "loss": 0.299,
      "step": 400
    },
    {
      "epoch": 79.01,
      "eval_accuracy": 0.8181818181818182,
      "eval_f1": 0.7575757575757577,
      "eval_loss": 0.46232300996780396,
      "eval_precision": 0.7272727272727273,
      "eval_recall": 0.8181818181818182,
      "eval_runtime": 20.4146,
      "eval_samples_per_second": 0.539,
      "eval_steps_per_second": 0.049,
      "step": 400
    },
    {
      "epoch": 80.01,
      "eval_accuracy": 0.9090909090909091,
      "eval_f1": 0.8787878787878788,
      "eval_loss": 0.4730096161365509,
      "eval_precision": 0.8636363636363636,
      "eval_recall": 0.9090909090909091,
      "eval_runtime": 20.8428,
      "eval_samples_per_second": 0.528,
      "eval_steps_per_second": 0.048,
      "step": 405
    },
    {
      "epoch": 81.01,
      "grad_norm": 3.940727949142456,
      "learning_rate": 3.62962962962963e-05,
      "loss": 0.1008,
      "step": 410
    },
    {
      "epoch": 81.01,
      "eval_accuracy": 0.8181818181818182,
      "eval_f1": 0.7878787878787878,
      "eval_loss": 0.582288920879364,
      "eval_precision": 0.7727272727272727,
      "eval_recall": 0.8181818181818182,
      "eval_runtime": 20.3701,
      "eval_samples_per_second": 0.54,
      "eval_steps_per_second": 0.049,
      "step": 410
    },
    {
      "epoch": 82.01,
      "eval_accuracy": 0.8181818181818182,
      "eval_f1": 0.8181818181818182,
      "eval_loss": 0.5622988939285278,
      "eval_precision": 0.8181818181818182,
      "eval_recall": 0.8181818181818182,
      "eval_runtime": 20.3908,
      "eval_samples_per_second": 0.539,
      "eval_steps_per_second": 0.049,
      "step": 415
    },
    {
      "epoch": 83.01,
      "grad_norm": 0.6912495493888855,
      "learning_rate": 3.555555555555556e-05,
      "loss": 0.1681,
      "step": 420
    },
    {
      "epoch": 83.01,
      "eval_accuracy": 0.9090909090909091,
      "eval_f1": 0.8787878787878788,
      "eval_loss": 0.44556429982185364,
      "eval_precision": 0.8636363636363636,
      "eval_recall": 0.9090909090909091,
      "eval_runtime": 20.9116,
      "eval_samples_per_second": 0.526,
      "eval_steps_per_second": 0.048,
      "step": 420
    },
    {
      "epoch": 84.01,
      "eval_accuracy": 0.9090909090909091,
      "eval_f1": 0.8787878787878788,
      "eval_loss": 0.3800961375236511,
      "eval_precision": 0.8636363636363636,
      "eval_recall": 0.9090909090909091,
      "eval_runtime": 20.9045,
      "eval_samples_per_second": 0.526,
      "eval_steps_per_second": 0.048,
      "step": 425
    },
    {
      "epoch": 85.01,
      "grad_norm": 1.7112609148025513,
      "learning_rate": 3.481481481481482e-05,
      "loss": 0.1318,
      "step": 430
    },
    {
      "epoch": 85.01,
      "eval_accuracy": 0.9090909090909091,
      "eval_f1": 0.8787878787878788,
      "eval_loss": 0.4264180362224579,
      "eval_precision": 0.8636363636363636,
      "eval_recall": 0.9090909090909091,
      "eval_runtime": 20.4232,
      "eval_samples_per_second": 0.539,
      "eval_steps_per_second": 0.049,
      "step": 430
    },
    {
      "epoch": 86.01,
      "eval_accuracy": 0.8181818181818182,
      "eval_f1": 0.8181818181818182,
      "eval_loss": 0.3410065472126007,
      "eval_precision": 0.8181818181818182,
      "eval_recall": 0.8181818181818182,
      "eval_runtime": 21.0143,
      "eval_samples_per_second": 0.523,
      "eval_steps_per_second": 0.048,
      "step": 435
    },
    {
      "epoch": 87.01,
      "grad_norm": 8.220293998718262,
      "learning_rate": 3.4074074074074077e-05,
      "loss": 0.1515,
      "step": 440
    },
    {
      "epoch": 87.01,
      "eval_accuracy": 0.7272727272727273,
      "eval_f1": 0.6818181818181818,
      "eval_loss": 0.6253748536109924,
      "eval_precision": 0.6666666666666666,
      "eval_recall": 0.7272727272727273,
      "eval_runtime": 20.6136,
      "eval_samples_per_second": 0.534,
      "eval_steps_per_second": 0.049,
      "step": 440
    },
    {
      "epoch": 88.01,
      "eval_accuracy": 0.7272727272727273,
      "eval_f1": 0.6818181818181818,
      "eval_loss": 0.5382604002952576,
      "eval_precision": 0.6666666666666666,
      "eval_recall": 0.7272727272727273,
      "eval_runtime": 20.2231,
      "eval_samples_per_second": 0.544,
      "eval_steps_per_second": 0.049,
      "step": 445
    },
    {
      "epoch": 89.01,
      "grad_norm": 7.24671745300293,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 0.1379,
      "step": 450
    },
    {
      "epoch": 89.01,
      "eval_accuracy": 0.9090909090909091,
      "eval_f1": 0.8787878787878788,
      "eval_loss": 0.26343244314193726,
      "eval_precision": 0.8636363636363636,
      "eval_recall": 0.9090909090909091,
      "eval_runtime": 22.4656,
      "eval_samples_per_second": 0.49,
      "eval_steps_per_second": 0.045,
      "step": 450
    }
  ],
  "logging_steps": 10,
  "max_steps": 900,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9223372036854775807,
  "save_steps": 500,
  "total_flos": 8.859993293767311e+18,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}