{
  "best_metric": 0.7887323943661972,
  "best_model_checkpoint": "./model_out/checkpoint-480",
  "epoch": 15.737704918032787,
  "eval_steps": 60,
  "global_step": 480,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.32786885245901637,
      "grad_norm": 3.083059549331665,
      "learning_rate": 9.833333333333333e-05,
      "loss": 1.3429,
      "step": 10
    },
    {
      "epoch": 0.6557377049180327,
      "grad_norm": 5.27333927154541,
      "learning_rate": 9.666666666666667e-05,
      "loss": 0.9792,
      "step": 20
    },
    {
      "epoch": 0.9836065573770492,
      "grad_norm": 6.412991046905518,
      "learning_rate": 9.5e-05,
      "loss": 0.8173,
      "step": 30
    },
    {
      "epoch": 1.3114754098360657,
      "grad_norm": 4.924794673919678,
      "learning_rate": 9.333333333333334e-05,
      "loss": 0.5787,
      "step": 40
    },
    {
      "epoch": 1.639344262295082,
      "grad_norm": 7.327770233154297,
      "learning_rate": 9.166666666666667e-05,
      "loss": 0.3744,
      "step": 50
    },
    {
      "epoch": 1.9672131147540983,
      "grad_norm": 8.928845405578613,
      "learning_rate": 9e-05,
      "loss": 0.4815,
      "step": 60
    },
    {
      "epoch": 1.9672131147540983,
      "eval_accuracy": 0.7897310513447433,
      "eval_f1": 0.6728971962616822,
      "eval_loss": 0.5841665267944336,
      "eval_precision": 0.6666666666666666,
      "eval_recall": 0.6792452830188679,
      "eval_runtime": 0.9794,
      "eval_samples_per_second": 61.26,
      "eval_steps_per_second": 8.168,
      "step": 60
    },
    {
      "epoch": 2.2950819672131146,
      "grad_norm": 5.491725444793701,
      "learning_rate": 8.833333333333333e-05,
      "loss": 0.2658,
      "step": 70
    },
    {
      "epoch": 2.6229508196721314,
      "grad_norm": 5.113138675689697,
      "learning_rate": 8.666666666666667e-05,
      "loss": 0.2402,
      "step": 80
    },
    {
      "epoch": 2.9508196721311473,
      "grad_norm": 6.449514865875244,
      "learning_rate": 8.5e-05,
      "loss": 0.1607,
      "step": 90
    },
    {
      "epoch": 3.278688524590164,
      "grad_norm": 5.699795246124268,
      "learning_rate": 8.333333333333334e-05,
      "loss": 0.0852,
      "step": 100
    },
    {
      "epoch": 3.6065573770491803,
      "grad_norm": 3.382863998413086,
      "learning_rate": 8.166666666666667e-05,
      "loss": 0.1513,
      "step": 110
    },
    {
      "epoch": 3.9344262295081966,
      "grad_norm": 2.5515336990356445,
      "learning_rate": 8e-05,
      "loss": 0.2612,
      "step": 120
    },
    {
      "epoch": 3.9344262295081966,
      "eval_accuracy": 0.8166259168704156,
      "eval_f1": 0.7009345794392523,
      "eval_loss": 0.7145981192588806,
      "eval_precision": 0.6944444444444444,
      "eval_recall": 0.7075471698113207,
      "eval_runtime": 0.9765,
      "eval_samples_per_second": 61.442,
      "eval_steps_per_second": 8.192,
      "step": 120
    },
    {
      "epoch": 4.262295081967213,
      "grad_norm": 3.4330668449401855,
      "learning_rate": 7.833333333333333e-05,
      "loss": 0.0857,
      "step": 130
    },
    {
      "epoch": 4.590163934426229,
      "grad_norm": 0.7570230960845947,
      "learning_rate": 7.666666666666667e-05,
      "loss": 0.0339,
      "step": 140
    },
    {
      "epoch": 4.918032786885246,
      "grad_norm": 13.973723411560059,
      "learning_rate": 7.500000000000001e-05,
      "loss": 0.0939,
      "step": 150
    },
    {
      "epoch": 5.245901639344262,
      "grad_norm": 6.072541236877441,
      "learning_rate": 7.333333333333333e-05,
      "loss": 0.0753,
      "step": 160
    },
    {
      "epoch": 5.573770491803279,
      "grad_norm": 6.689024925231934,
      "learning_rate": 7.166666666666667e-05,
      "loss": 0.0497,
      "step": 170
    },
    {
      "epoch": 5.901639344262295,
      "grad_norm": 2.9444985389709473,
      "learning_rate": 7e-05,
      "loss": 0.0535,
      "step": 180
    },
    {
      "epoch": 5.901639344262295,
      "eval_accuracy": 0.8166259168704156,
      "eval_f1": 0.6697247706422018,
      "eval_loss": 0.7874237298965454,
      "eval_precision": 0.6517857142857143,
      "eval_recall": 0.6886792452830188,
      "eval_runtime": 0.9781,
      "eval_samples_per_second": 61.346,
      "eval_steps_per_second": 8.179,
      "step": 180
    },
    {
      "epoch": 6.229508196721311,
      "grad_norm": 1.6225782632827759,
      "learning_rate": 6.833333333333333e-05,
      "loss": 0.0258,
      "step": 190
    },
    {
      "epoch": 6.557377049180328,
      "grad_norm": 0.07454058527946472,
      "learning_rate": 6.666666666666667e-05,
      "loss": 0.0273,
      "step": 200
    },
    {
      "epoch": 6.885245901639344,
      "grad_norm": 5.249931812286377,
      "learning_rate": 6.500000000000001e-05,
      "loss": 0.023,
      "step": 210
    },
    {
      "epoch": 7.213114754098361,
      "grad_norm": 0.14014603197574615,
      "learning_rate": 6.333333333333333e-05,
      "loss": 0.0162,
      "step": 220
    },
    {
      "epoch": 7.540983606557377,
      "grad_norm": 0.024733418598771095,
      "learning_rate": 6.166666666666667e-05,
      "loss": 0.0099,
      "step": 230
    },
    {
      "epoch": 7.868852459016393,
      "grad_norm": 1.4897031784057617,
      "learning_rate": 6e-05,
      "loss": 0.0141,
      "step": 240
    },
    {
      "epoch": 7.868852459016393,
      "eval_accuracy": 0.8557457212713936,
      "eval_f1": 0.780269058295964,
      "eval_loss": 0.8452157378196716,
      "eval_precision": 0.7435897435897436,
      "eval_recall": 0.8207547169811321,
      "eval_runtime": 0.9777,
      "eval_samples_per_second": 61.371,
      "eval_steps_per_second": 8.183,
      "step": 240
    },
    {
      "epoch": 8.19672131147541,
      "grad_norm": 6.675982475280762,
      "learning_rate": 5.833333333333334e-05,
      "loss": 0.0201,
      "step": 250
    },
    {
      "epoch": 8.524590163934427,
      "grad_norm": 0.21795310080051422,
      "learning_rate": 5.666666666666667e-05,
      "loss": 0.0098,
      "step": 260
    },
    {
      "epoch": 8.852459016393443,
      "grad_norm": 1.0106713771820068,
      "learning_rate": 5.500000000000001e-05,
      "loss": 0.0162,
      "step": 270
    },
    {
      "epoch": 9.180327868852459,
      "grad_norm": 1.9501103162765503,
      "learning_rate": 5.333333333333333e-05,
      "loss": 0.0166,
      "step": 280
    },
    {
      "epoch": 9.508196721311476,
      "grad_norm": 0.09368854761123657,
      "learning_rate": 5.166666666666667e-05,
      "loss": 0.0178,
      "step": 290
    },
    {
      "epoch": 9.836065573770492,
      "grad_norm": 5.705907821655273,
      "learning_rate": 5e-05,
      "loss": 0.0102,
      "step": 300
    },
    {
      "epoch": 9.836065573770492,
      "eval_accuracy": 0.8068459657701712,
      "eval_f1": 0.7296137339055794,
      "eval_loss": 1.098240613937378,
      "eval_precision": 0.6692913385826772,
      "eval_recall": 0.8018867924528302,
      "eval_runtime": 0.9802,
      "eval_samples_per_second": 61.211,
      "eval_steps_per_second": 8.161,
      "step": 300
    },
    {
      "epoch": 10.163934426229508,
      "grad_norm": 0.016423122957348824,
      "learning_rate": 4.8333333333333334e-05,
      "loss": 0.0048,
      "step": 310
    },
    {
      "epoch": 10.491803278688524,
      "grad_norm": 0.013950828462839127,
      "learning_rate": 4.666666666666667e-05,
      "loss": 0.0023,
      "step": 320
    },
    {
      "epoch": 10.819672131147541,
      "grad_norm": 0.0043854909017682076,
      "learning_rate": 4.5e-05,
      "loss": 0.0049,
      "step": 330
    },
    {
      "epoch": 11.147540983606557,
      "grad_norm": 0.5211315751075745,
      "learning_rate": 4.3333333333333334e-05,
      "loss": 0.0218,
      "step": 340
    },
    {
      "epoch": 11.475409836065573,
      "grad_norm": 0.0140004837885499,
      "learning_rate": 4.166666666666667e-05,
      "loss": 0.0085,
      "step": 350
    },
    {
      "epoch": 11.80327868852459,
      "grad_norm": 1.8254053592681885,
      "learning_rate": 4e-05,
      "loss": 0.0099,
      "step": 360
    },
    {
      "epoch": 11.80327868852459,
      "eval_accuracy": 0.80440097799511,
      "eval_f1": 0.7203791469194313,
      "eval_loss": 1.3635536432266235,
      "eval_precision": 0.7238095238095238,
      "eval_recall": 0.7169811320754716,
      "eval_runtime": 0.9823,
      "eval_samples_per_second": 61.081,
      "eval_steps_per_second": 8.144,
      "step": 360
    },
    {
      "epoch": 12.131147540983607,
      "grad_norm": 0.1549140065908432,
      "learning_rate": 3.8333333333333334e-05,
      "loss": 0.0004,
      "step": 370
    },
    {
      "epoch": 12.459016393442623,
      "grad_norm": 4.475136756896973,
      "learning_rate": 3.6666666666666666e-05,
      "loss": 0.0026,
      "step": 380
    },
    {
      "epoch": 12.78688524590164,
      "grad_norm": 11.31729507446289,
      "learning_rate": 3.5e-05,
      "loss": 0.0024,
      "step": 390
    },
    {
      "epoch": 13.114754098360656,
      "grad_norm": 0.003214745782315731,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 0.0012,
      "step": 400
    },
    {
      "epoch": 13.442622950819672,
      "grad_norm": 0.9436277747154236,
      "learning_rate": 3.1666666666666666e-05,
      "loss": 0.0145,
      "step": 410
    },
    {
      "epoch": 13.770491803278688,
      "grad_norm": 5.284872055053711,
      "learning_rate": 3e-05,
      "loss": 0.0087,
      "step": 420
    },
    {
      "epoch": 13.770491803278688,
      "eval_accuracy": 0.8166259168704156,
      "eval_f1": 0.7751196172248804,
      "eval_loss": 1.346622347831726,
      "eval_precision": 0.7864077669902912,
      "eval_recall": 0.7641509433962265,
      "eval_runtime": 0.9819,
      "eval_samples_per_second": 61.108,
      "eval_steps_per_second": 8.148,
      "step": 420
    },
    {
      "epoch": 14.098360655737705,
      "grad_norm": 0.003362592775374651,
      "learning_rate": 2.8333333333333335e-05,
      "loss": 0.0011,
      "step": 430
    },
    {
      "epoch": 14.426229508196721,
      "grad_norm": 0.011366785503923893,
      "learning_rate": 2.6666666666666667e-05,
      "loss": 0.0003,
      "step": 440
    },
    {
      "epoch": 14.754098360655737,
      "grad_norm": 0.003698177170008421,
      "learning_rate": 2.5e-05,
      "loss": 0.0002,
      "step": 450
    },
    {
      "epoch": 15.081967213114755,
      "grad_norm": 0.0021139541640877724,
      "learning_rate": 2.3333333333333336e-05,
      "loss": 0.0047,
      "step": 460
    },
    {
      "epoch": 15.40983606557377,
      "grad_norm": 0.0036122933961451054,
      "learning_rate": 2.1666666666666667e-05,
      "loss": 0.0002,
      "step": 470
    },
    {
      "epoch": 15.737704918032787,
      "grad_norm": 0.010879104025661945,
      "learning_rate": 2e-05,
      "loss": 0.0002,
      "step": 480
    },
    {
      "epoch": 15.737704918032787,
      "eval_accuracy": 0.8288508557457213,
      "eval_f1": 0.7887323943661972,
      "eval_loss": 1.271989107131958,
      "eval_precision": 0.7850467289719626,
      "eval_recall": 0.7924528301886793,
      "eval_runtime": 0.9809,
      "eval_samples_per_second": 61.168,
      "eval_steps_per_second": 8.156,
      "step": 480
    }
  ],
  "logging_steps": 10,
  "max_steps": 600,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 20,
  "save_steps": 60,
  "total_flos": 995567645030400.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}