trainer_state.json

{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 20.0,
  "eval_steps": 500,
  "global_step": 100,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 4.8252272605896,
      "learning_rate": 4.800000000000001e-06,
      "loss": 0.7486,
      "step": 5
    },
    {
      "epoch": 2.0,
      "grad_norm": 3.033998489379883,
      "learning_rate": 4.5500000000000005e-06,
      "loss": 0.312,
      "step": 10
    },
    {
      "epoch": 3.0,
      "grad_norm": 1.8606371879577637,
      "learning_rate": 4.3e-06,
      "loss": 0.1901,
      "step": 15
    },
    {
      "epoch": 4.0,
      "grad_norm": 1.6860159635543823,
      "learning_rate": 4.05e-06,
      "loss": 0.1365,
      "step": 20
    },
    {
      "epoch": 5.0,
      "grad_norm": 1.3219197988510132,
      "learning_rate": 3.8000000000000005e-06,
      "loss": 0.1032,
      "step": 25
    },
    {
      "epoch": 6.0,
      "grad_norm": 1.1769614219665527,
      "learning_rate": 3.5500000000000003e-06,
      "loss": 0.0779,
      "step": 30
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.9814537763595581,
      "learning_rate": 3.3000000000000006e-06,
      "loss": 0.0607,
      "step": 35
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.9276354908943176,
      "learning_rate": 3.05e-06,
      "loss": 0.0476,
      "step": 40
    },
    {
      "epoch": 9.0,
      "grad_norm": 0.8008437156677246,
      "learning_rate": 2.8000000000000003e-06,
      "loss": 0.0385,
      "step": 45
    },
    {
      "epoch": 10.0,
      "grad_norm": 0.8980335593223572,
      "learning_rate": 2.55e-06,
      "loss": 0.0326,
      "step": 50
    },
    {
      "epoch": 11.0,
      "grad_norm": 0.8411991000175476,
      "learning_rate": 2.3000000000000004e-06,
      "loss": 0.028,
      "step": 55
    },
    {
      "epoch": 12.0,
      "grad_norm": 1.4981303215026855,
      "learning_rate": 2.05e-06,
      "loss": 0.0256,
      "step": 60
    },
    {
      "epoch": 13.0,
      "grad_norm": 0.7783492803573608,
      "learning_rate": 1.8000000000000001e-06,
      "loss": 0.023,
      "step": 65
    },
    {
      "epoch": 14.0,
      "grad_norm": 0.78788822889328,
      "learning_rate": 1.5500000000000002e-06,
      "loss": 0.0219,
      "step": 70
    },
    {
      "epoch": 15.0,
      "grad_norm": 0.5979446768760681,
      "learning_rate": 1.3e-06,
      "loss": 0.0205,
      "step": 75
    },
    {
      "epoch": 16.0,
      "grad_norm": 0.5709032416343689,
      "learning_rate": 1.0500000000000001e-06,
      "loss": 0.0202,
      "step": 80
    },
    {
      "epoch": 17.0,
      "grad_norm": 0.5151193141937256,
      "learning_rate": 8.000000000000001e-07,
      "loss": 0.0194,
      "step": 85
    },
    {
      "epoch": 18.0,
      "grad_norm": 0.597676694393158,
      "learning_rate": 5.5e-07,
      "loss": 0.0191,
      "step": 90
    },
    {
      "epoch": 19.0,
      "grad_norm": 0.7428526282310486,
      "learning_rate": 3.0000000000000004e-07,
      "loss": 0.0188,
      "step": 95
    },
    {
      "epoch": 20.0,
      "grad_norm": 0.6364104151725769,
      "learning_rate": 5.0000000000000004e-08,
      "loss": 0.0186,
      "step": 100
    }
  ],
  "logging_steps": 5,
  "max_steps": 100,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 20,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1759140800102400.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}