trainer_state.json

{
  "best_global_step": 34,
  "best_metric": 0.6140967607498169,
  "best_model_checkpoint": "distilbert_rand_50_v2_cola/checkpoint-34",
  "epoch": 6.0,
  "eval_steps": 500,
  "global_step": 204,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 1.4710649251937866,
      "learning_rate": 4.9e-05,
      "loss": 0.6138,
      "step": 34
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6912751793861389,
      "eval_loss": 0.6140967607498169,
      "eval_matthews_correlation": 0.0,
      "eval_runtime": 0.4482,
      "eval_samples_per_second": 2327.247,
      "eval_steps_per_second": 11.157,
      "step": 34
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.9415231943130493,
      "learning_rate": 4.8e-05,
      "loss": 0.5901,
      "step": 68
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6912751793861389,
      "eval_loss": 0.6174531579017639,
      "eval_matthews_correlation": 0.025690322354642595,
      "eval_runtime": 0.4747,
      "eval_samples_per_second": 2197.37,
      "eval_steps_per_second": 10.534,
      "step": 68
    },
    {
      "epoch": 3.0,
      "grad_norm": 2.3130075931549072,
      "learning_rate": 4.7e-05,
      "loss": 0.5446,
      "step": 102
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6826462149620056,
      "eval_loss": 0.6304358243942261,
      "eval_matthews_correlation": 0.08323045697852056,
      "eval_runtime": 0.433,
      "eval_samples_per_second": 2408.97,
      "eval_steps_per_second": 11.548,
      "step": 102
    },
    {
      "epoch": 4.0,
      "grad_norm": 4.770295143127441,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.4862,
      "step": 136
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6366251111030579,
      "eval_loss": 0.7370460629463196,
      "eval_matthews_correlation": 0.08989469784450668,
      "eval_runtime": 0.4535,
      "eval_samples_per_second": 2300.128,
      "eval_steps_per_second": 11.026,
      "step": 136
    },
    {
      "epoch": 5.0,
      "grad_norm": 2.3743627071380615,
      "learning_rate": 4.5e-05,
      "loss": 0.4254,
      "step": 170
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.6634707450866699,
      "eval_loss": 0.7217957973480225,
      "eval_matthews_correlation": 0.10864609205493897,
      "eval_runtime": 0.4405,
      "eval_samples_per_second": 2367.684,
      "eval_steps_per_second": 11.35,
      "step": 170
    },
    {
      "epoch": 6.0,
      "grad_norm": 2.990626096725464,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.372,
      "step": 204
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.6078619360923767,
      "eval_loss": 0.8227274417877197,
      "eval_matthews_correlation": 0.07252846630353921,
      "eval_runtime": 0.4985,
      "eval_samples_per_second": 2092.27,
      "eval_steps_per_second": 10.03,
      "step": 204
    },
    {
      "epoch": 6.0,
      "step": 204,
      "total_flos": 3398186177722368.0,
      "train_loss": 0.5053384958529005,
      "train_runtime": 58.532,
      "train_samples_per_second": 7304.554,
      "train_steps_per_second": 29.044
    }
  ],
  "logging_steps": 1,
  "max_steps": 1700,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 50,
  "save_steps": 500,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 5,
        "early_stopping_threshold": 0.0
      },
      "attributes": {
        "early_stopping_patience_counter": 5
      }
    },
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 3398186177722368.0,
  "train_batch_size": 256,
  "trial_name": null,
  "trial_params": null
}