run-0/checkpoint-3207/trainer_state.json

{
  "best_metric": 0.4756138930857138,
  "best_model_checkpoint": "distilbert-base-uncased-finetuned-cola/run-0/checkpoint-3207",
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 3207,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.4677268475210477,
      "grad_norm": 8.62888240814209,
      "learning_rate": 5.373752855939329e-06,
      "loss": 0.5755,
      "step": 500
    },
    {
      "epoch": 0.9354536950420954,
      "grad_norm": 20.624385833740234,
      "learning_rate": 4.381186757686775e-06,
      "loss": 0.5165,
      "step": 1000
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.5140782594680786,
      "eval_matthews_correlation": 0.4102877158206736,
      "eval_runtime": 2.0196,
      "eval_samples_per_second": 516.439,
      "eval_steps_per_second": 32.68,
      "step": 1069
    },
    {
      "epoch": 1.4031805425631432,
      "grad_norm": 10.336002349853516,
      "learning_rate": 3.38862065943422e-06,
      "loss": 0.4425,
      "step": 1500
    },
    {
      "epoch": 1.8709073900841908,
      "grad_norm": 11.466729164123535,
      "learning_rate": 2.396054561181666e-06,
      "loss": 0.4078,
      "step": 2000
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.5298678874969482,
      "eval_matthews_correlation": 0.44455763518973207,
      "eval_runtime": 2.0041,
      "eval_samples_per_second": 520.439,
      "eval_steps_per_second": 32.933,
      "step": 2138
    },
    {
      "epoch": 2.3386342376052385,
      "grad_norm": 22.138774871826172,
      "learning_rate": 1.4034884629291118e-06,
      "loss": 0.3583,
      "step": 2500
    },
    {
      "epoch": 2.8063610851262863,
      "grad_norm": 7.407865524291992,
      "learning_rate": 4.1092236467655754e-07,
      "loss": 0.3541,
      "step": 3000
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.5446668267250061,
      "eval_matthews_correlation": 0.4756138930857138,
      "eval_runtime": 2.3169,
      "eval_samples_per_second": 450.175,
      "eval_steps_per_second": 28.487,
      "step": 3207
    }
  ],
  "logging_steps": 500,
  "max_steps": 3207,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 112738035468672.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": {
    "learning_rate": 6.366318954191884e-06,
    "num_train_epochs": 3,
    "per_device_train_batch_size": 8,
    "seed": 21
  }
}