trainer_state.json

{
  "best_metric": 0.8410995602607727,
  "best_model_checkpoint": "out/checkpoint-3",
  "epoch": 0.0014160962945480293,
  "eval_steps": 1,
  "global_step": 3,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "grad_norm": 2.342660903930664,
      "learning_rate": 0.0002,
      "loss": 1.4012,
      "step": 1
    },
    {
      "epoch": 0.0,
      "eval_loss": 1.0993317365646362,
      "eval_runtime": 5834.6353,
      "eval_samples_per_second": 0.366,
      "eval_steps_per_second": 0.366,
      "step": 1
    },
    {
      "epoch": 0.0,
      "grad_norm": 1.581579566001892,
      "learning_rate": 0.0002,
      "loss": 1.0882,
      "step": 2
    },
    {
      "epoch": 0.0,
      "eval_loss": 0.958124577999115,
      "eval_runtime": 5835.2952,
      "eval_samples_per_second": 0.366,
      "eval_steps_per_second": 0.366,
      "step": 2
    },
    {
      "epoch": 0.0,
      "grad_norm": 1.2699007987976074,
      "learning_rate": 0.0002,
      "loss": 0.962,
      "step": 3
    },
    {
      "epoch": 0.0,
      "eval_loss": 0.8410995602607727,
      "eval_runtime": 5835.1886,
      "eval_samples_per_second": 0.366,
      "eval_steps_per_second": 0.366,
      "step": 3
    }
  ],
  "logging_steps": 1,
  "max_steps": 10,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 1,
  "total_flos": 525800935784448.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}