trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 4.444444444444445,
  "eval_steps": 1000,
  "global_step": 800,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.5555555555555556,
      "grad_norm": 18.646621704101562,
      "learning_rate": 1.7777777777777777e-05,
      "loss": 12.9379,
      "step": 100
    },
    {
      "epoch": 1.1111111111111112,
      "grad_norm": 15.787500381469727,
      "learning_rate": 1.555555555555556e-05,
      "loss": 12.4243,
      "step": 200
    },
    {
      "epoch": 1.6666666666666665,
      "grad_norm": 24.62742805480957,
      "learning_rate": 1.3333333333333333e-05,
      "loss": 12.1655,
      "step": 300
    },
    {
      "epoch": 2.2222222222222223,
      "grad_norm": 22.622655868530273,
      "learning_rate": 1.1111111111111113e-05,
      "loss": 11.9433,
      "step": 400
    },
    {
      "epoch": 2.7777777777777777,
      "grad_norm": 24.53895378112793,
      "learning_rate": 8.888888888888888e-06,
      "loss": 11.9653,
      "step": 500
    },
    {
      "epoch": 3.3333333333333335,
      "grad_norm": 29.335464477539062,
      "learning_rate": 6.666666666666667e-06,
      "loss": 11.8621,
      "step": 600
    },
    {
      "epoch": 3.888888888888889,
      "grad_norm": 23.47711753845215,
      "learning_rate": 4.444444444444444e-06,
      "loss": 11.5167,
      "step": 700
    },
    {
      "epoch": 4.444444444444445,
      "grad_norm": 21.372766494750977,
      "learning_rate": 2.222222222222222e-06,
      "loss": 11.3985,
      "step": 800
    }
  ],
  "logging_steps": 100,
  "max_steps": 900,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 100,
  "total_flos": 1007165894833152.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}