trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 6729,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.22291573785109228,
      "grad_norm": 1.3647773265838623,
      "learning_rate": 5e-05,
      "loss": 0.7791,
      "step": 500
    },
    {
      "epoch": 0.44583147570218457,
      "grad_norm": 0.9170280694961548,
      "learning_rate": 4.598651468935624e-05,
      "loss": 0.1977,
      "step": 1000
    },
    {
      "epoch": 0.6687472135532768,
      "grad_norm": 0.8765759468078613,
      "learning_rate": 4.1973029378712475e-05,
      "loss": 0.1697,
      "step": 1500
    },
    {
      "epoch": 0.8916629514043691,
      "grad_norm": 0.6906009912490845,
      "learning_rate": 3.7959544068068715e-05,
      "loss": 0.1548,
      "step": 2000
    },
    {
      "epoch": 1.1145786892554614,
      "grad_norm": 0.9499293565750122,
      "learning_rate": 3.394605875742495e-05,
      "loss": 0.1323,
      "step": 2500
    },
    {
      "epoch": 1.3374944271065536,
      "grad_norm": 0.822642982006073,
      "learning_rate": 2.9932573446781187e-05,
      "loss": 0.1174,
      "step": 3000
    },
    {
      "epoch": 1.560410164957646,
      "grad_norm": 0.757134735584259,
      "learning_rate": 2.591908813613742e-05,
      "loss": 0.1101,
      "step": 3500
    },
    {
      "epoch": 1.7833259028087383,
      "grad_norm": 0.6683522462844849,
      "learning_rate": 2.190560282549366e-05,
      "loss": 0.1051,
      "step": 4000
    },
    {
      "epoch": 2.0062416406598307,
      "grad_norm": 0.7783832550048828,
      "learning_rate": 1.7892117514849896e-05,
      "loss": 0.1017,
      "step": 4500
    },
    {
      "epoch": 2.229157378510923,
      "grad_norm": 0.6604039669036865,
      "learning_rate": 1.3878632204206132e-05,
      "loss": 0.0858,
      "step": 5000
    },
    {
      "epoch": 2.452073116362015,
      "grad_norm": 0.7059195637702942,
      "learning_rate": 9.86514689356237e-06,
      "loss": 0.0835,
      "step": 5500
    },
    {
      "epoch": 2.6749888542131073,
      "grad_norm": 0.6601874232292175,
      "learning_rate": 5.851661582918607e-06,
      "loss": 0.0824,
      "step": 6000
    },
    {
      "epoch": 2.8979045920642,
      "grad_norm": 0.6698402762413025,
      "learning_rate": 1.8381762722748435e-06,
      "loss": 0.0816,
      "step": 6500
    }
  ],
  "logging_steps": 500,
  "max_steps": 6729,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 10000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 7031368581120000.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}