run-8/checkpoint-1070/trainer_state.json

{
  "best_metric": 0.4867950830885312,
  "best_model_checkpoint": "distilbert-base-uncased-finetuned-cola/run-8/checkpoint-1070",
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 1070,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.9345794392523364,
      "grad_norm": 5.401679039001465,
      "learning_rate": 3.394171042589578e-05,
      "loss": 0.5368,
      "step": 500
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.4921383559703827,
      "eval_matthews_correlation": 0.4184145751988745,
      "eval_runtime": 0.341,
      "eval_samples_per_second": 3059.08,
      "eval_steps_per_second": 193.576,
      "step": 535
    },
    {
      "epoch": 1.8691588785046729,
      "grad_norm": 8.301382064819336,
      "learning_rate": 4.168280227741587e-06,
      "loss": 0.2872,
      "step": 1000
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.5593526363372803,
      "eval_matthews_correlation": 0.4867950830885312,
      "eval_runtime": 0.3536,
      "eval_samples_per_second": 2949.8,
      "eval_steps_per_second": 186.66,
      "step": 1070
    }
  ],
  "logging_steps": 500,
  "max_steps": 1070,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 500,
  "total_flos": 88497536416176.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": {
    "learning_rate": 6.371514062404997e-05,
    "num_train_epochs": 2,
    "per_device_train_batch_size": 16,
    "seed": 18
  }
}