checkpoint-100/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.5384615384615383,
  "eval_steps": 25,
  "global_step": 100,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.38,
      "learning_rate": 2.29933110367893e-05,
      "loss": 1.9196,
      "step": 25
    },
    {
      "epoch": 0.38,
      "eval_loss": 1.4835141897201538,
      "eval_runtime": 4.4399,
      "eval_samples_per_second": 4.505,
      "eval_steps_per_second": 0.676,
      "step": 25
    },
    {
      "epoch": 0.77,
      "learning_rate": 2.090301003344482e-05,
      "loss": 0.9398,
      "step": 50
    },
    {
      "epoch": 0.77,
      "eval_loss": 1.353461503982544,
      "eval_runtime": 4.4353,
      "eval_samples_per_second": 4.509,
      "eval_steps_per_second": 0.676,
      "step": 50
    },
    {
      "epoch": 1.15,
      "learning_rate": 1.8812709030100337e-05,
      "loss": 0.7795,
      "step": 75
    },
    {
      "epoch": 1.15,
      "eval_loss": 1.4102122783660889,
      "eval_runtime": 4.4451,
      "eval_samples_per_second": 4.499,
      "eval_steps_per_second": 0.675,
      "step": 75
    },
    {
      "epoch": 1.54,
      "learning_rate": 1.6722408026755853e-05,
      "loss": 0.6915,
      "step": 100
    },
    {
      "epoch": 1.54,
      "eval_loss": 1.3842780590057373,
      "eval_runtime": 4.44,
      "eval_samples_per_second": 4.505,
      "eval_steps_per_second": 0.676,
      "step": 100
    }
  ],
  "logging_steps": 25,
  "max_steps": 300,
  "num_train_epochs": 5,
  "save_steps": 25,
  "total_flos": 552629855846400.0,
  "trial_name": null,
  "trial_params": null
}