checkpoint-10/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.4,
  "eval_steps": 5,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 3.1814,
      "step": 1
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.00018571428571428572,
      "loss": 3.5839,
      "step": 2
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.00017142857142857143,
      "loss": 3.7658,
      "step": 3
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.00015714285714285716,
      "loss": 3.6873,
      "step": 4
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.00014285714285714287,
      "loss": 3.6731,
      "step": 5
    },
    {
      "epoch": 0.2,
      "eval_loss": 3.442739963531494,
      "eval_runtime": 53.5168,
      "eval_samples_per_second": 0.187,
      "eval_steps_per_second": 0.037,
      "step": 5
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.00012857142857142858,
      "loss": 3.9097,
      "step": 6
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.00011428571428571428,
      "loss": 3.0121,
      "step": 7
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.0001,
      "loss": 3.2549,
      "step": 8
    },
    {
      "epoch": 0.36,
      "learning_rate": 8.571428571428571e-05,
      "loss": 3.4586,
      "step": 9
    },
    {
      "epoch": 0.4,
      "learning_rate": 7.142857142857143e-05,
      "loss": 3.3894,
      "step": 10
    },
    {
      "epoch": 0.4,
      "eval_loss": 3.212744951248169,
      "eval_runtime": 59.9058,
      "eval_samples_per_second": 0.167,
      "eval_steps_per_second": 0.033,
      "step": 10
    }
  ],
  "logging_steps": 1,
  "max_steps": 15,
  "num_train_epochs": 1,
  "save_steps": 5,
  "total_flos": 35005697851392.0,
  "trial_name": null,
  "trial_params": null
}