checkpoint-100/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.06420545746388442,
  "global_step": 100,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "learning_rate": 0.00019800000000000002,
      "loss": 35.1383,
      "step": 10
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.000196,
      "loss": 27.6025,
      "step": 20
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.000194,
      "loss": 3.9383,
      "step": 30
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.000194,
      "loss": 18.6562,
      "step": 40
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.000194,
      "loss": 4.7225,
      "step": 50
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.000194,
      "loss": 26.334,
      "step": 60
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.000194,
      "loss": 49.6182,
      "step": 70
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.000194,
      "loss": 29.6643,
      "step": 80
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.000194,
      "loss": 14.5124,
      "step": 90
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.000194,
      "loss": 37.2745,
      "step": 100
    }
  ],
  "max_steps": 100,
  "num_train_epochs": 1,
  "total_flos": 1.3390613878800384e+16,
  "trial_name": null,
  "trial_params": null
}