trainer_state.json

{
  "best_global_step": 34,
  "best_metric": 0.6183428764343262,
  "best_model_checkpoint": "distilbert_rand_10_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": 1.044641375541687,
      "learning_rate": 4.9e-05,
      "loss": 0.6159,
      "step": 34
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6912751793861389,
      "eval_loss": 0.6183428764343262,
      "eval_matthews_correlation": 0.0,
      "eval_runtime": 0.4462,
      "eval_samples_per_second": 2337.555,
      "eval_steps_per_second": 11.206,
      "step": 34
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.3998757600784302,
      "learning_rate": 4.8e-05,
      "loss": 0.5927,
      "step": 68
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6912751793861389,
      "eval_loss": 0.6238136887550354,
      "eval_matthews_correlation": 0.06620236880686593,
      "eval_runtime": 0.4558,
      "eval_samples_per_second": 2288.039,
      "eval_steps_per_second": 10.969,
      "step": 68
    },
    {
      "epoch": 3.0,
      "grad_norm": 2.483593702316284,
      "learning_rate": 4.7e-05,
      "loss": 0.5523,
      "step": 102
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6816874146461487,
      "eval_loss": 0.6280689835548401,
      "eval_matthews_correlation": 0.04141506615334042,
      "eval_runtime": 0.4512,
      "eval_samples_per_second": 2311.479,
      "eval_steps_per_second": 11.081,
      "step": 102
    },
    {
      "epoch": 4.0,
      "grad_norm": 3.1675617694854736,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.4975,
      "step": 136
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6529242396354675,
      "eval_loss": 0.701947808265686,
      "eval_matthews_correlation": 0.07393732833556993,
      "eval_runtime": 0.4603,
      "eval_samples_per_second": 2265.857,
      "eval_steps_per_second": 10.862,
      "step": 136
    },
    {
      "epoch": 5.0,
      "grad_norm": 2.3917527198791504,
      "learning_rate": 4.5e-05,
      "loss": 0.4393,
      "step": 170
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.6500479578971863,
      "eval_loss": 0.7124446034431458,
      "eval_matthews_correlation": 0.09743296097822274,
      "eval_runtime": 0.435,
      "eval_samples_per_second": 2397.922,
      "eval_steps_per_second": 11.495,
      "step": 170
    },
    {
      "epoch": 6.0,
      "grad_norm": 3.3322689533233643,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.3833,
      "step": 204
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.6308724880218506,
      "eval_loss": 0.8296592235565186,
      "eval_matthews_correlation": 0.05821172078977795,
      "eval_runtime": 0.4375,
      "eval_samples_per_second": 2383.732,
      "eval_steps_per_second": 11.427,
      "step": 204
    },
    {
      "epoch": 6.0,
      "step": 204,
      "total_flos": 3398186177722368.0,
      "train_loss": 0.5134991758009967,
      "train_runtime": 57.633,
      "train_samples_per_second": 7418.489,
      "train_steps_per_second": 29.497
    }
  ],
  "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
}