trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9574468085106383,
  "eval_steps": 500,
  "global_step": 90,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.10638297872340426,
      "grad_norm": 37.91831970214844,
      "learning_rate": 8.936170212765958e-05,
      "loss": 0.4663,
      "step": 10
    },
    {
      "epoch": 0.2127659574468085,
      "grad_norm": 4.650197982788086,
      "learning_rate": 7.872340425531916e-05,
      "loss": 0.2238,
      "step": 20
    },
    {
      "epoch": 0.3191489361702128,
      "grad_norm": 3.5010766983032227,
      "learning_rate": 6.808510638297873e-05,
      "loss": 0.1096,
      "step": 30
    },
    {
      "epoch": 0.425531914893617,
      "grad_norm": 2.2222304344177246,
      "learning_rate": 5.744680851063831e-05,
      "loss": 0.0581,
      "step": 40
    },
    {
      "epoch": 0.5319148936170213,
      "grad_norm": 5.912722110748291,
      "learning_rate": 4.680851063829788e-05,
      "loss": 0.0451,
      "step": 50
    },
    {
      "epoch": 0.6382978723404256,
      "grad_norm": 0.5471988916397095,
      "learning_rate": 3.617021276595745e-05,
      "loss": 0.03,
      "step": 60
    },
    {
      "epoch": 0.7446808510638298,
      "grad_norm": 5.3497138023376465,
      "learning_rate": 2.5531914893617022e-05,
      "loss": 0.0238,
      "step": 70
    },
    {
      "epoch": 0.851063829787234,
      "grad_norm": 0.6585015654563904,
      "learning_rate": 1.4893617021276596e-05,
      "loss": 0.0167,
      "step": 80
    },
    {
      "epoch": 0.9574468085106383,
      "grad_norm": 0.7444632649421692,
      "learning_rate": 4.255319148936171e-06,
      "loss": 0.0133,
      "step": 90
    }
  ],
  "logging_steps": 10,
  "max_steps": 94,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 10,
  "total_flos": 0.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}