trainer_state.json

{
  "best_global_step": 264,
  "best_metric": 0.4457773268222809,
  "best_model_checkpoint": "distilbert_km_10_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": 7.7640228271484375,
      "learning_rate": 4.9e-05,
      "loss": 0.3956,
      "step": 264
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8142201834862385,
      "eval_loss": 0.4457773268222809,
      "eval_runtime": 0.3877,
      "eval_samples_per_second": 2249.319,
      "eval_steps_per_second": 10.318,
      "step": 264
    },
    {
      "epoch": 2.0,
      "grad_norm": 4.566969871520996,
      "learning_rate": 4.8e-05,
      "loss": 0.2194,
      "step": 528
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8153669724770642,
      "eval_loss": 0.48116862773895264,
      "eval_runtime": 0.367,
      "eval_samples_per_second": 2375.723,
      "eval_steps_per_second": 10.898,
      "step": 528
    },
    {
      "epoch": 3.0,
      "grad_norm": 3.1546401977539062,
      "learning_rate": 4.7e-05,
      "loss": 0.1562,
      "step": 792
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8027522935779816,
      "eval_loss": 0.5327261686325073,
      "eval_runtime": 0.3764,
      "eval_samples_per_second": 2316.88,
      "eval_steps_per_second": 10.628,
      "step": 792
    },
    {
      "epoch": 4.0,
      "grad_norm": 5.845862865447998,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.1129,
      "step": 1056
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.7970183486238532,
      "eval_loss": 0.6541174054145813,
      "eval_runtime": 0.3724,
      "eval_samples_per_second": 2341.518,
      "eval_steps_per_second": 10.741,
      "step": 1056
    },
    {
      "epoch": 5.0,
      "grad_norm": 2.4241580963134766,
      "learning_rate": 4.5e-05,
      "loss": 0.083,
      "step": 1320
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8061926605504587,
      "eval_loss": 0.66544508934021,
      "eval_runtime": 0.366,
      "eval_samples_per_second": 2382.632,
      "eval_steps_per_second": 10.93,
      "step": 1320
    },
    {
      "epoch": 6.0,
      "grad_norm": 7.1404523849487305,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.0636,
      "step": 1584
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8004587155963303,
      "eval_loss": 0.7401790618896484,
      "eval_runtime": 0.3888,
      "eval_samples_per_second": 2242.604,
      "eval_steps_per_second": 10.287,
      "step": 1584
    },
    {
      "epoch": 6.0,
      "step": 1584,
      "total_flos": 2.676464049624883e+16,
      "train_loss": 0.17180385974922566,
      "train_runtime": 379.1464,
      "train_samples_per_second": 8881.661,
      "train_steps_per_second": 34.815
    }
  ],
  "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
}