trainer_state.json

{
  "best_global_step": 264,
  "best_metric": 0.44564706087112427,
  "best_model_checkpoint": "distilbert_rand_20_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": 9.27222728729248,
      "learning_rate": 4.9e-05,
      "loss": 0.3907,
      "step": 264
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8027522935779816,
      "eval_loss": 0.44564706087112427,
      "eval_runtime": 0.3862,
      "eval_samples_per_second": 2258.182,
      "eval_steps_per_second": 10.359,
      "step": 264
    },
    {
      "epoch": 2.0,
      "grad_norm": 9.175657272338867,
      "learning_rate": 4.8e-05,
      "loss": 0.2199,
      "step": 528
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8245412844036697,
      "eval_loss": 0.4951857030391693,
      "eval_runtime": 0.367,
      "eval_samples_per_second": 2376.05,
      "eval_steps_per_second": 10.899,
      "step": 528
    },
    {
      "epoch": 3.0,
      "grad_norm": 4.057657241821289,
      "learning_rate": 4.7e-05,
      "loss": 0.1616,
      "step": 792
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8188073394495413,
      "eval_loss": 0.49606284499168396,
      "eval_runtime": 0.3676,
      "eval_samples_per_second": 2372.449,
      "eval_steps_per_second": 10.883,
      "step": 792
    },
    {
      "epoch": 4.0,
      "grad_norm": 5.760934829711914,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.1207,
      "step": 1056
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8061926605504587,
      "eval_loss": 0.6331028342247009,
      "eval_runtime": 0.3649,
      "eval_samples_per_second": 2389.787,
      "eval_steps_per_second": 10.962,
      "step": 1056
    },
    {
      "epoch": 5.0,
      "grad_norm": 14.224475860595703,
      "learning_rate": 4.5e-05,
      "loss": 0.0916,
      "step": 1320
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.7970183486238532,
      "eval_loss": 0.6100755333900452,
      "eval_runtime": 0.3734,
      "eval_samples_per_second": 2335.007,
      "eval_steps_per_second": 10.711,
      "step": 1320
    },
    {
      "epoch": 6.0,
      "grad_norm": 7.731550693511963,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.0728,
      "step": 1584
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8004587155963303,
      "eval_loss": 0.7173300981521606,
      "eval_runtime": 0.4232,
      "eval_samples_per_second": 2060.26,
      "eval_steps_per_second": 9.451,
      "step": 1584
    },
    {
      "epoch": 6.0,
      "step": 1584,
      "total_flos": 2.676464049624883e+16,
      "train_loss": 0.17622880863420892,
      "train_runtime": 376.2004,
      "train_samples_per_second": 8951.213,
      "train_steps_per_second": 35.088
    }
  ],
  "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
}