trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 10.0,
  "eval_steps": 500,
  "global_step": 19420,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 1.0474146604537964,
      "learning_rate": 9.932086511308698e-05,
      "loss": 0.7626,
      "step": 1942
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.34481489658355713,
      "learning_rate": 9.398178969222977e-05,
      "loss": 0.2322,
      "step": 3884
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.41964319348335266,
      "learning_rate": 8.387660491043735e-05,
      "loss": 0.1471,
      "step": 5826
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.8369811773300171,
      "learning_rate": 7.01081600148828e-05,
      "loss": 0.1128,
      "step": 7768
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.27606743574142456,
      "learning_rate": 5.4154422280282916e-05,
      "loss": 0.096,
      "step": 9710
    },
    {
      "epoch": 6.0,
      "grad_norm": 1.0203839540481567,
      "learning_rate": 3.775048842911087e-05,
      "loss": 0.0836,
      "step": 11652
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.27634188532829285,
      "learning_rate": 2.2681109856932215e-05,
      "loss": 0.0769,
      "step": 13594
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.1679644137620926,
      "learning_rate": 1.0563901955241806e-05,
      "loss": 0.0709,
      "step": 15536
    },
    {
      "epoch": 9.0,
      "grad_norm": 0.0655561313033104,
      "learning_rate": 2.720207190809809e-06,
      "loss": 0.0667,
      "step": 17478
    },
    {
      "epoch": 10.0,
      "grad_norm": 0.24803902208805084,
      "learning_rate": 1.1598826660241813e-11,
      "loss": 0.0636,
      "step": 19420
    }
  ],
  "logging_steps": 500,
  "max_steps": 19420,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "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": 1.0777248659423232e+17,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}