trainer_state.json

{
  "best_global_step": 264,
  "best_metric": 0.4527788758277893,
  "best_model_checkpoint": "distilbert_rand_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": 7.377885341644287,
      "learning_rate": 4.9e-05,
      "loss": 0.3883,
      "step": 264
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8188073394495413,
      "eval_loss": 0.4527788758277893,
      "eval_runtime": 0.3745,
      "eval_samples_per_second": 2328.485,
      "eval_steps_per_second": 10.681,
      "step": 264
    },
    {
      "epoch": 2.0,
      "grad_norm": 6.007850170135498,
      "learning_rate": 4.8e-05,
      "loss": 0.2203,
      "step": 528
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8188073394495413,
      "eval_loss": 0.47409456968307495,
      "eval_runtime": 0.3723,
      "eval_samples_per_second": 2341.946,
      "eval_steps_per_second": 10.743,
      "step": 528
    },
    {
      "epoch": 3.0,
      "grad_norm": 4.742299556732178,
      "learning_rate": 4.7e-05,
      "loss": 0.1626,
      "step": 792
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8084862385321101,
      "eval_loss": 0.514573872089386,
      "eval_runtime": 0.375,
      "eval_samples_per_second": 2325.316,
      "eval_steps_per_second": 10.667,
      "step": 792
    },
    {
      "epoch": 4.0,
      "grad_norm": 5.826347827911377,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.123,
      "step": 1056
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8268348623853211,
      "eval_loss": 0.5761303901672363,
      "eval_runtime": 0.3797,
      "eval_samples_per_second": 2296.576,
      "eval_steps_per_second": 10.535,
      "step": 1056
    },
    {
      "epoch": 5.0,
      "grad_norm": 5.340493679046631,
      "learning_rate": 4.5e-05,
      "loss": 0.0957,
      "step": 1320
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8096330275229358,
      "eval_loss": 0.6311283111572266,
      "eval_runtime": 0.3748,
      "eval_samples_per_second": 2326.742,
      "eval_steps_per_second": 10.673,
      "step": 1320
    },
    {
      "epoch": 6.0,
      "grad_norm": 3.8428404331207275,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.0747,
      "step": 1584
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8027522935779816,
      "eval_loss": 0.6833432912826538,
      "eval_runtime": 0.3896,
      "eval_samples_per_second": 2237.986,
      "eval_steps_per_second": 10.266,
      "step": 1584
    },
    {
      "epoch": 6.0,
      "step": 1584,
      "total_flos": 2.676464049624883e+16,
      "train_loss": 0.1774315195854264,
      "train_runtime": 377.5983,
      "train_samples_per_second": 8918.074,
      "train_steps_per_second": 34.958
    }
  ],
  "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
}