trainer_state.json

{
  "best_global_step": 264,
  "best_metric": 0.4374975562095642,
  "best_model_checkpoint": "distilbert_rand_100_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": 8.47227954864502,
      "learning_rate": 4.9e-05,
      "loss": 0.3983,
      "step": 264
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8119266055045872,
      "eval_loss": 0.4374975562095642,
      "eval_runtime": 0.3628,
      "eval_samples_per_second": 2403.246,
      "eval_steps_per_second": 11.024,
      "step": 264
    },
    {
      "epoch": 2.0,
      "grad_norm": 6.604650974273682,
      "learning_rate": 4.8e-05,
      "loss": 0.2241,
      "step": 528
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8165137614678899,
      "eval_loss": 0.4724898636341095,
      "eval_runtime": 0.4012,
      "eval_samples_per_second": 2173.336,
      "eval_steps_per_second": 9.969,
      "step": 528
    },
    {
      "epoch": 3.0,
      "grad_norm": 4.221500873565674,
      "learning_rate": 4.7e-05,
      "loss": 0.1648,
      "step": 792
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8107798165137615,
      "eval_loss": 0.4910937547683716,
      "eval_runtime": 0.373,
      "eval_samples_per_second": 2337.889,
      "eval_steps_per_second": 10.724,
      "step": 792
    },
    {
      "epoch": 4.0,
      "grad_norm": 5.404686450958252,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.1242,
      "step": 1056
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.7981651376146789,
      "eval_loss": 0.6093521118164062,
      "eval_runtime": 0.3663,
      "eval_samples_per_second": 2380.424,
      "eval_steps_per_second": 10.919,
      "step": 1056
    },
    {
      "epoch": 5.0,
      "grad_norm": 2.832387685775757,
      "learning_rate": 4.5e-05,
      "loss": 0.092,
      "step": 1320
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8027522935779816,
      "eval_loss": 0.6336649656295776,
      "eval_runtime": 0.3676,
      "eval_samples_per_second": 2372.293,
      "eval_steps_per_second": 10.882,
      "step": 1320
    },
    {
      "epoch": 6.0,
      "grad_norm": 8.062549591064453,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.0717,
      "step": 1584
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.7981651376146789,
      "eval_loss": 0.8973585963249207,
      "eval_runtime": 0.3766,
      "eval_samples_per_second": 2315.435,
      "eval_steps_per_second": 10.621,
      "step": 1584
    },
    {
      "epoch": 6.0,
      "step": 1584,
      "total_flos": 2.676464049624883e+16,
      "train_loss": 0.17917960701566754,
      "train_runtime": 377.5481,
      "train_samples_per_second": 8919.26,
      "train_steps_per_second": 34.962
    }
  ],
  "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
}