{
  "best_metric": 0.6372231085109245,
  "best_model_checkpoint": "roberta-base-finetuned-cola/checkpoint-1070",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 2675,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.93,
      "grad_norm": 16.3082275390625,
      "learning_rate": 1.6261682242990654e-05,
      "loss": 0.4777,
      "step": 500
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.43241721391677856,
      "eval_matthews_correlation": 0.5493333540758192,
      "eval_runtime": 1.2622,
      "eval_samples_per_second": 826.355,
      "eval_steps_per_second": 52.291,
      "step": 535
    },
    {
      "epoch": 1.87,
      "grad_norm": 23.89521598815918,
      "learning_rate": 1.2523364485981309e-05,
      "loss": 0.3138,
      "step": 1000
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.4074053466320038,
      "eval_matthews_correlation": 0.6372231085109245,
      "eval_runtime": 1.315,
      "eval_samples_per_second": 793.178,
      "eval_steps_per_second": 50.191,
      "step": 1070
    },
    {
      "epoch": 2.8,
      "grad_norm": 18.608760833740234,
      "learning_rate": 8.785046728971963e-06,
      "loss": 0.2262,
      "step": 1500
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.6061288714408875,
      "eval_matthews_correlation": 0.576660981486724,
      "eval_runtime": 1.3277,
      "eval_samples_per_second": 785.578,
      "eval_steps_per_second": 49.711,
      "step": 1605
    },
    {
      "epoch": 3.74,
      "grad_norm": 2.4630796909332275,
      "learning_rate": 5.046728971962617e-06,
      "loss": 0.1786,
      "step": 2000
    },
    {
      "epoch": 4.0,
      "eval_loss": 0.884406328201294,
      "eval_matthews_correlation": 0.5855361143791362,
      "eval_runtime": 1.2996,
      "eval_samples_per_second": 802.571,
      "eval_steps_per_second": 50.786,
      "step": 2140
    },
    {
      "epoch": 4.67,
      "grad_norm": 0.05961179360747337,
      "learning_rate": 1.308411214953271e-06,
      "loss": 0.1253,
      "step": 2500
    },
    {
      "epoch": 5.0,
      "eval_loss": 0.9215266704559326,
      "eval_matthews_correlation": 0.6158197390467447,
      "eval_runtime": 1.2999,
      "eval_samples_per_second": 802.375,
      "eval_steps_per_second": 50.773,
      "step": 2675
    }
  ],
  "logging_steps": 500,
  "max_steps": 2675,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "total_flos": 449687113048260.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}