trainer_state.json

{
  "best_global_step": 34,
  "best_metric": 0.6141772866249084,
  "best_model_checkpoint": "distilbert_rand_20_v1_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": 1.278737187385559,
      "learning_rate": 4.9e-05,
      "loss": 0.6142,
      "step": 34
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6912751793861389,
      "eval_loss": 0.6141772866249084,
      "eval_matthews_correlation": 0.0,
      "eval_runtime": 0.4369,
      "eval_samples_per_second": 2387.061,
      "eval_steps_per_second": 11.443,
      "step": 34
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.4612891674041748,
      "learning_rate": 4.8e-05,
      "loss": 0.5913,
      "step": 68
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6883988380432129,
      "eval_loss": 0.6215904355049133,
      "eval_matthews_correlation": 0.012271814729887918,
      "eval_runtime": 0.4799,
      "eval_samples_per_second": 2173.54,
      "eval_steps_per_second": 10.42,
      "step": 68
    },
    {
      "epoch": 3.0,
      "grad_norm": 2.1359140872955322,
      "learning_rate": 4.7e-05,
      "loss": 0.55,
      "step": 102
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6797698736190796,
      "eval_loss": 0.6157238483428955,
      "eval_matthews_correlation": 0.053976421365281024,
      "eval_runtime": 0.446,
      "eval_samples_per_second": 2338.342,
      "eval_steps_per_second": 11.21,
      "step": 102
    },
    {
      "epoch": 4.0,
      "grad_norm": 3.3808982372283936,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.5024,
      "step": 136
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.638542652130127,
      "eval_loss": 0.6957675218582153,
      "eval_matthews_correlation": 0.09639180875142503,
      "eval_runtime": 0.4381,
      "eval_samples_per_second": 2380.982,
      "eval_steps_per_second": 11.414,
      "step": 136
    },
    {
      "epoch": 5.0,
      "grad_norm": 2.731954336166382,
      "learning_rate": 4.5e-05,
      "loss": 0.4358,
      "step": 170
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.6586769223213196,
      "eval_loss": 0.7210583090782166,
      "eval_matthews_correlation": 0.10400282407668013,
      "eval_runtime": 0.4745,
      "eval_samples_per_second": 2198.196,
      "eval_steps_per_second": 10.538,
      "step": 170
    },
    {
      "epoch": 6.0,
      "grad_norm": 2.567495107650757,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.3857,
      "step": 204
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.6548417806625366,
      "eval_loss": 0.8117982149124146,
      "eval_matthews_correlation": 0.0932488887627139,
      "eval_runtime": 0.4391,
      "eval_samples_per_second": 2375.487,
      "eval_steps_per_second": 11.388,
      "step": 204
    },
    {
      "epoch": 6.0,
      "step": 204,
      "total_flos": 3398186177722368.0,
      "train_loss": 0.51321387758442,
      "train_runtime": 57.9585,
      "train_samples_per_second": 7376.831,
      "train_steps_per_second": 29.331
    }
  ],
  "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
}