checkpoint-400/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 14.006666666666666,
  "global_step": 400,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "learning_rate": 4.9999999999999996e-06,
      "loss": 11.0688,
      "step": 1
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.00025,
      "loss": 8.3635,
      "step": 50
    },
    {
      "epoch": 3.01,
      "learning_rate": 0.0005,
      "loss": 6.4596,
      "step": 100
    },
    {
      "epoch": 5.01,
      "learning_rate": 0.0005833333333333333,
      "loss": 5.8337,
      "step": 150
    },
    {
      "epoch": 7.0,
      "learning_rate": 0.0005555555555555556,
      "loss": 5.4472,
      "step": 200
    },
    {
      "epoch": 8.02,
      "learning_rate": 0.0005277777777777777,
      "loss": 5.1974,
      "step": 250
    },
    {
      "epoch": 10.02,
      "learning_rate": 0.0005,
      "loss": 5.1091,
      "step": 300
    },
    {
      "epoch": 12.01,
      "learning_rate": 0.00047222222222222224,
      "loss": 4.9724,
      "step": 350
    },
    {
      "epoch": 14.01,
      "learning_rate": 0.00044444444444444436,
      "loss": 4.8054,
      "step": 400
    }
  ],
  "max_steps": 1200,
  "num_train_epochs": 9223372036854775807,
  "total_flos": 1.08554297278464e+17,
  "trial_name": null,
  "trial_params": null
}