{
  "best_metric": 0.5129449367523193,
  "best_model_checkpoint": "output_pipe/tf3/origin/checkpoint-800",
  "epoch": 4.0,
  "eval_steps": 200,
  "global_step": 1708,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.234192037470726,
      "grad_norm": 3.9260871410369873,
      "learning_rate": 2.9095295536791317e-05,
      "loss": 0.6316,
      "step": 100
    },
    {
      "epoch": 0.468384074941452,
      "grad_norm": 3.6461851596832275,
      "learning_rate": 2.7285886610373945e-05,
      "loss": 0.5777,
      "step": 200
    },
    {
      "epoch": 0.468384074941452,
      "eval_accuracy": 0.723,
      "eval_f1": 0.722999722999723,
      "eval_loss": 0.5350773930549622,
      "eval_matthews_correlation": 0.44777297076054284,
      "eval_precision": 0.7238707946386023,
      "eval_recall": 0.7239021772216794,
      "eval_runtime": 0.2568,
      "eval_samples_per_second": 3893.706,
      "eval_steps_per_second": 62.299,
      "step": 200
    },
    {
      "epoch": 0.702576112412178,
      "grad_norm": 2.2741026878356934,
      "learning_rate": 2.5476477683956577e-05,
      "loss": 0.5575,
      "step": 300
    },
    {
      "epoch": 0.936768149882904,
      "grad_norm": 1.7103469371795654,
      "learning_rate": 2.3667068757539205e-05,
      "loss": 0.5257,
      "step": 400
    },
    {
      "epoch": 0.936768149882904,
      "eval_accuracy": 0.738,
      "eval_f1": 0.734550214588796,
      "eval_loss": 0.5157595872879028,
      "eval_matthews_correlation": 0.4782909026279838,
      "eval_precision": 0.7435211641969566,
      "eval_recall": 0.7348483634790689,
      "eval_runtime": 0.2548,
      "eval_samples_per_second": 3924.007,
      "eval_steps_per_second": 62.784,
      "step": 400
    },
    {
      "epoch": 1.17096018735363,
      "grad_norm": 5.01882791519165,
      "learning_rate": 2.1857659831121837e-05,
      "loss": 0.4602,
      "step": 500
    },
    {
      "epoch": 1.405152224824356,
      "grad_norm": 8.720273971557617,
      "learning_rate": 2.006634499396864e-05,
      "loss": 0.4395,
      "step": 600
    },
    {
      "epoch": 1.405152224824356,
      "eval_accuracy": 0.723,
      "eval_f1": 0.7220323947661811,
      "eval_loss": 0.5580781698226929,
      "eval_matthews_correlation": 0.4572919910512899,
      "eval_precision": 0.7312564786476067,
      "eval_recall": 0.7260649802143578,
      "eval_runtime": 0.2548,
      "eval_samples_per_second": 3925.211,
      "eval_steps_per_second": 62.803,
      "step": 600
    },
    {
      "epoch": 1.639344262295082,
      "grad_norm": 3.7575089931488037,
      "learning_rate": 1.8256936067551267e-05,
      "loss": 0.4317,
      "step": 700
    },
    {
      "epoch": 1.8735362997658078,
      "grad_norm": 3.8411054611206055,
      "learning_rate": 1.64475271411339e-05,
      "loss": 0.421,
      "step": 800
    },
    {
      "epoch": 1.8735362997658078,
      "eval_accuracy": 0.759,
      "eval_f1": 0.7589129675812969,
      "eval_loss": 0.5129449367523193,
      "eval_matthews_correlation": 0.5182649804477735,
      "eval_precision": 0.7589650358601434,
      "eval_recall": 0.7593000528685176,
      "eval_runtime": 0.2554,
      "eval_samples_per_second": 3914.909,
      "eval_steps_per_second": 62.639,
      "step": 800
    },
    {
      "epoch": 2.107728337236534,
      "grad_norm": 5.369962215423584,
      "learning_rate": 1.4638118214716525e-05,
      "loss": 0.3326,
      "step": 900
    },
    {
      "epoch": 2.34192037470726,
      "grad_norm": 3.9363551139831543,
      "learning_rate": 1.2828709288299155e-05,
      "loss": 0.2206,
      "step": 1000
    },
    {
      "epoch": 2.34192037470726,
      "eval_accuracy": 0.746,
      "eval_f1": 0.7459908556708041,
      "eval_loss": 0.6723874807357788,
      "eval_matthews_correlation": 0.49461837601059455,
      "eval_precision": 0.7474340881390724,
      "eval_recall": 0.7471843509187908,
      "eval_runtime": 0.255,
      "eval_samples_per_second": 3921.156,
      "eval_steps_per_second": 62.739,
      "step": 1000
    },
    {
      "epoch": 2.576112412177986,
      "grad_norm": 4.995495319366455,
      "learning_rate": 1.1019300361881785e-05,
      "loss": 0.1944,
      "step": 1100
    },
    {
      "epoch": 2.810304449648712,
      "grad_norm": 2.590538263320923,
      "learning_rate": 9.209891435464416e-06,
      "loss": 0.2089,
      "step": 1200
    },
    {
      "epoch": 2.810304449648712,
      "eval_accuracy": 0.744,
      "eval_f1": 0.743990783668212,
      "eval_loss": 0.7384678721427917,
      "eval_matthews_correlation": 0.48932203716456163,
      "eval_precision": 0.7445728739754098,
      "eval_recall": 0.7447491949566638,
      "eval_runtime": 0.2549,
      "eval_samples_per_second": 3923.28,
      "eval_steps_per_second": 62.772,
      "step": 1200
    },
    {
      "epoch": 3.0444964871194378,
      "grad_norm": 8.24785041809082,
      "learning_rate": 7.400482509047045e-06,
      "loss": 0.1618,
      "step": 1300
    },
    {
      "epoch": 3.278688524590164,
      "grad_norm": 5.278331279754639,
      "learning_rate": 5.591073582629674e-06,
      "loss": 0.0524,
      "step": 1400
    },
    {
      "epoch": 3.278688524590164,
      "eval_accuracy": 0.738,
      "eval_f1": 0.7379486379330349,
      "eval_loss": 1.1256791353225708,
      "eval_matthews_correlation": 0.4766121867752247,
      "eval_precision": 0.7381592421915002,
      "eval_recall": 0.738453035133533,
      "eval_runtime": 0.2551,
      "eval_samples_per_second": 3919.39,
      "eval_steps_per_second": 62.71,
      "step": 1400
    },
    {
      "epoch": 3.51288056206089,
      "grad_norm": 3.5331642627716064,
      "learning_rate": 3.781664656212304e-06,
      "loss": 0.0452,
      "step": 1500
    },
    {
      "epoch": 3.747072599531616,
      "grad_norm": 1.4989087581634521,
      "learning_rate": 1.9722557297949336e-06,
      "loss": 0.0414,
      "step": 1600
    },
    {
      "epoch": 3.747072599531616,
      "eval_accuracy": 0.733,
      "eval_f1": 0.7328586822429065,
      "eval_loss": 1.189754843711853,
      "eval_matthews_correlation": 0.4659653271721009,
      "eval_precision": 0.7328432843284328,
      "eval_recall": 0.7331221262756533,
      "eval_runtime": 0.2546,
      "eval_samples_per_second": 3927.942,
      "eval_steps_per_second": 62.847,
      "step": 1600
    },
    {
      "epoch": 3.981264637002342,
      "grad_norm": 5.889491558074951,
      "learning_rate": 1.6284680337756335e-07,
      "loss": 0.0444,
      "step": 1700
    },
    {
      "epoch": 4.0,
      "step": 1708,
      "total_flos": 7349802265996560.0,
      "train_loss": 0.3131386968950617,
      "train_runtime": 147.5142,
      "train_samples_per_second": 740.105,
      "train_steps_per_second": 11.579
    }
  ],
  "logging_steps": 100,
  "max_steps": 1708,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 200,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 7349802265996560.0,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}