trainer_state.json

{
  "best_global_step": 264,
  "best_metric": 0.43098530173301697,
  "best_model_checkpoint": "distilbert_km_5_v2_sst2/checkpoint-264",
  "epoch": 6.0,
  "eval_steps": 500,
  "global_step": 1584,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 9.637043952941895,
      "learning_rate": 4.9e-05,
      "loss": 0.3757,
      "step": 264
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8291284403669725,
      "eval_loss": 0.43098530173301697,
      "eval_runtime": 0.3693,
      "eval_samples_per_second": 2361.412,
      "eval_steps_per_second": 10.832,
      "step": 264
    },
    {
      "epoch": 2.0,
      "grad_norm": 8.912264823913574,
      "learning_rate": 4.8e-05,
      "loss": 0.2117,
      "step": 528
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8142201834862385,
      "eval_loss": 0.4856370985507965,
      "eval_runtime": 0.3644,
      "eval_samples_per_second": 2392.997,
      "eval_steps_per_second": 10.977,
      "step": 528
    },
    {
      "epoch": 3.0,
      "grad_norm": 7.051660537719727,
      "learning_rate": 4.7e-05,
      "loss": 0.1487,
      "step": 792
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8130733944954128,
      "eval_loss": 0.5372852087020874,
      "eval_runtime": 0.3648,
      "eval_samples_per_second": 2390.673,
      "eval_steps_per_second": 10.966,
      "step": 792
    },
    {
      "epoch": 4.0,
      "grad_norm": 7.2147698402404785,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.1058,
      "step": 1056
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.823394495412844,
      "eval_loss": 0.6082008481025696,
      "eval_runtime": 0.364,
      "eval_samples_per_second": 2395.362,
      "eval_steps_per_second": 10.988,
      "step": 1056
    },
    {
      "epoch": 5.0,
      "grad_norm": 12.811993598937988,
      "learning_rate": 4.5e-05,
      "loss": 0.0796,
      "step": 1320
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8119266055045872,
      "eval_loss": 0.6212149858474731,
      "eval_runtime": 0.3764,
      "eval_samples_per_second": 2316.403,
      "eval_steps_per_second": 10.626,
      "step": 1320
    },
    {
      "epoch": 6.0,
      "grad_norm": 7.478900909423828,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.0585,
      "step": 1584
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.823394495412844,
      "eval_loss": 0.7494320869445801,
      "eval_runtime": 0.3689,
      "eval_samples_per_second": 2363.978,
      "eval_steps_per_second": 10.844,
      "step": 1584
    },
    {
      "epoch": 6.0,
      "step": 1584,
      "total_flos": 2.676464049624883e+16,
      "train_loss": 0.16332613879984076,
      "train_runtime": 376.4488,
      "train_samples_per_second": 8945.308,
      "train_steps_per_second": 35.065
    }
  ],
  "logging_steps": 1,
  "max_steps": 13200,
  "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": 2.676464049624883e+16,
  "train_batch_size": 256,
  "trial_name": null,
  "trial_params": null
}