trainer_state.json

{
  "best_global_step": 264,
  "best_metric": 0.45564746856689453,
  "best_model_checkpoint": "distilbert_rand_10_v1_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": 8.012245178222656,
      "learning_rate": 4.9e-05,
      "loss": 0.3899,
      "step": 264
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8176605504587156,
      "eval_loss": 0.45564746856689453,
      "eval_runtime": 0.3948,
      "eval_samples_per_second": 2208.78,
      "eval_steps_per_second": 10.132,
      "step": 264
    },
    {
      "epoch": 2.0,
      "grad_norm": 8.25125789642334,
      "learning_rate": 4.8e-05,
      "loss": 0.2228,
      "step": 528
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8142201834862385,
      "eval_loss": 0.5009775161743164,
      "eval_runtime": 0.3772,
      "eval_samples_per_second": 2311.55,
      "eval_steps_per_second": 10.603,
      "step": 528
    },
    {
      "epoch": 3.0,
      "grad_norm": 5.213468551635742,
      "learning_rate": 4.7e-05,
      "loss": 0.1643,
      "step": 792
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7993119266055045,
      "eval_loss": 0.5151464939117432,
      "eval_runtime": 0.3691,
      "eval_samples_per_second": 2362.51,
      "eval_steps_per_second": 10.837,
      "step": 792
    },
    {
      "epoch": 4.0,
      "grad_norm": 6.990106582641602,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.1223,
      "step": 1056
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8245412844036697,
      "eval_loss": 0.5864481329917908,
      "eval_runtime": 0.367,
      "eval_samples_per_second": 2376.044,
      "eval_steps_per_second": 10.899,
      "step": 1056
    },
    {
      "epoch": 5.0,
      "grad_norm": 4.5160441398620605,
      "learning_rate": 4.5e-05,
      "loss": 0.093,
      "step": 1320
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8188073394495413,
      "eval_loss": 0.58154296875,
      "eval_runtime": 0.3723,
      "eval_samples_per_second": 2342.49,
      "eval_steps_per_second": 10.745,
      "step": 1320
    },
    {
      "epoch": 6.0,
      "grad_norm": 6.444607257843018,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.0743,
      "step": 1584
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8188073394495413,
      "eval_loss": 0.628352165222168,
      "eval_runtime": 0.3697,
      "eval_samples_per_second": 2358.863,
      "eval_steps_per_second": 10.82,
      "step": 1584
    },
    {
      "epoch": 6.0,
      "step": 1584,
      "total_flos": 2.676464049624883e+16,
      "train_loss": 0.17776139095576124,
      "train_runtime": 377.1201,
      "train_samples_per_second": 8929.383,
      "train_steps_per_second": 35.002
    }
  ],
  "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
}