{
  "best_global_step": 465,
  "best_metric": 0.7806512878301866,
  "best_model_checkpoint": "/kaggle/working/transformer_output/checkpoint-465",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 775,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.3225806451612903,
      "grad_norm": 3.0797810554504395,
      "learning_rate": 1.2727272727272728e-05,
      "loss": 1.1631031799316407,
      "step": 50
    },
    {
      "epoch": 0.6451612903225806,
      "grad_norm": 3.368655204772949,
      "learning_rate": 1.9951016547387286e-05,
      "loss": 0.6313877105712891,
      "step": 100
    },
    {
      "epoch": 0.967741935483871,
      "grad_norm": 5.035543918609619,
      "learning_rate": 1.947950097086825e-05,
      "loss": 0.5252903747558594,
      "step": 150
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8930589184826473,
      "eval_f1": 0.900084594871237,
      "eval_hate_f1": 0.5027322404371585,
      "eval_hate_prec": 0.4125560538116592,
      "eval_hate_rec": 0.6433566433566433,
      "eval_loss": 0.2638593912124634,
      "eval_macro_f1": 0.7711109967956045,
      "eval_precision": 0.9107537030140248,
      "eval_recall": 0.8930589184826473,
      "eval_runtime": 4.8064,
      "eval_samples_per_second": 515.563,
      "eval_steps_per_second": 8.114,
      "step": 155
    },
    {
      "epoch": 1.2903225806451613,
      "grad_norm": 2.4748122692108154,
      "learning_rate": 1.8529927230890757e-05,
      "loss": 0.501370506286621,
      "step": 200
    },
    {
      "epoch": 1.6129032258064515,
      "grad_norm": 2.85589599609375,
      "learning_rate": 1.715018302529852e-05,
      "loss": 0.4371611404418945,
      "step": 250
    },
    {
      "epoch": 1.935483870967742,
      "grad_norm": 2.5343403816223145,
      "learning_rate": 1.5409849862969994e-05,
      "loss": 0.44288436889648436,
      "step": 300
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8930589184826473,
      "eval_f1": 0.9015939212097996,
      "eval_hate_f1": 0.49604221635883905,
      "eval_hate_prec": 0.3983050847457627,
      "eval_hate_rec": 0.6573426573426573,
      "eval_loss": 0.24266110360622406,
      "eval_macro_f1": 0.7724936604193823,
      "eval_precision": 0.9157473422011047,
      "eval_recall": 0.8930589184826473,
      "eval_runtime": 4.9035,
      "eval_samples_per_second": 505.358,
      "eval_steps_per_second": 7.954,
      "step": 310
    },
    {
      "epoch": 2.258064516129032,
      "grad_norm": 6.783071994781494,
      "learning_rate": 1.3396694017461708e-05,
      "loss": 0.4083076477050781,
      "step": 350
    },
    {
      "epoch": 2.5806451612903225,
      "grad_norm": 4.064752578735352,
      "learning_rate": 1.1212240409675825e-05,
      "loss": 0.36163642883300784,
      "step": 400
    },
    {
      "epoch": 2.903225806451613,
      "grad_norm": 3.120911121368408,
      "learning_rate": 8.966652631884506e-06,
      "loss": 0.3937630081176758,
      "step": 450
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8970944309927361,
      "eval_f1": 0.9050220867632944,
      "eval_hate_f1": 0.5053191489361702,
      "eval_hate_prec": 0.40772532188841204,
      "eval_hate_rec": 0.6643356643356644,
      "eval_loss": 0.23981328308582306,
      "eval_macro_f1": 0.7806512878301866,
      "eval_precision": 0.9177721181109427,
      "eval_recall": 0.8970944309927361,
      "eval_runtime": 4.8801,
      "eval_samples_per_second": 507.772,
      "eval_steps_per_second": 7.992,
      "step": 465
    },
    {
      "epoch": 3.225806451612903,
      "grad_norm": 2.958275556564331,
      "learning_rate": 6.773177317242257e-06,
      "loss": 0.3346284103393555,
      "step": 500
    },
    {
      "epoch": 3.5483870967741935,
      "grad_norm": 2.155022144317627,
      "learning_rate": 4.742433029292856e-06,
      "loss": 0.3073293113708496,
      "step": 550
    },
    {
      "epoch": 3.870967741935484,
      "grad_norm": 3.33176589012146,
      "learning_rate": 2.976831686947884e-06,
      "loss": 0.31076955795288086,
      "step": 600
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.887409200968523,
      "eval_f1": 0.8985370069312554,
      "eval_hate_f1": 0.47665847665847666,
      "eval_hate_prec": 0.36742424242424243,
      "eval_hate_rec": 0.6783216783216783,
      "eval_loss": 0.24611614644527435,
      "eval_macro_f1": 0.7688956887260704,
      "eval_precision": 0.9169174752269499,
      "eval_recall": 0.887409200968523,
      "eval_runtime": 4.927,
      "eval_samples_per_second": 502.941,
      "eval_steps_per_second": 7.916,
      "step": 620
    },
    {
      "epoch": 4.193548387096774,
      "grad_norm": 3.568920612335205,
      "learning_rate": 1.565413856553095e-06,
      "loss": 0.3023691749572754,
      "step": 650
    },
    {
      "epoch": 4.516129032258064,
      "grad_norm": 3.1700286865234375,
      "learning_rate": 5.79358371039106e-07,
      "loss": 0.2629020118713379,
      "step": 700
    },
    {
      "epoch": 4.838709677419355,
      "grad_norm": 3.6812124252319336,
      "learning_rate": 6.839272934511143e-08,
      "loss": 0.27848766326904295,
      "step": 750
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8857949959644875,
      "eval_f1": 0.8976954173118289,
      "eval_hate_f1": 0.472289156626506,
      "eval_hate_prec": 0.3602941176470588,
      "eval_hate_rec": 0.6853146853146853,
      "eval_loss": 0.2520979344844818,
      "eval_macro_f1": 0.7679642000169223,
      "eval_precision": 0.9173899334739389,
      "eval_recall": 0.8857949959644875,
      "eval_runtime": 4.8991,
      "eval_samples_per_second": 505.803,
      "eval_steps_per_second": 7.961,
      "step": 775
    }
  ],
  "logging_steps": 50,
  "max_steps": 775,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 2,
        "early_stopping_threshold": 0.0
      },
      "attributes": {
        "early_stopping_patience_counter": 2
      }
    },
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 2445228103066560.0,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}