trainer_state.json

{
  "best_metric": 1.0493139028549194,
  "best_model_checkpoint": "out/checkpoint-7",
  "epoch": 0.009992862241256246,
  "eval_steps": 1,
  "global_step": 7,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "grad_norm": 8.742387771606445,
      "learning_rate": 0.0002,
      "loss": 2.8568,
      "step": 1
    },
    {
      "epoch": 0.0,
      "eval_loss": 2.0093040466308594,
      "eval_runtime": 1917.976,
      "eval_samples_per_second": 0.365,
      "eval_steps_per_second": 0.365,
      "step": 1
    },
    {
      "epoch": 0.0,
      "grad_norm": 4.034029006958008,
      "learning_rate": 0.0002,
      "loss": 2.0005,
      "step": 2
    },
    {
      "epoch": 0.0,
      "eval_loss": 1.51231050491333,
      "eval_runtime": 1918.2314,
      "eval_samples_per_second": 0.365,
      "eval_steps_per_second": 0.365,
      "step": 2
    },
    {
      "epoch": 0.0,
      "grad_norm": 3.6795763969421387,
      "learning_rate": 0.0002,
      "loss": 1.5174,
      "step": 3
    },
    {
      "epoch": 0.0,
      "eval_loss": 1.2310951948165894,
      "eval_runtime": 1918.2692,
      "eval_samples_per_second": 0.365,
      "eval_steps_per_second": 0.365,
      "step": 3
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.265252709388733,
      "learning_rate": 0.0002,
      "loss": 1.1661,
      "step": 4
    },
    {
      "epoch": 0.01,
      "eval_loss": 1.1966514587402344,
      "eval_runtime": 1918.1163,
      "eval_samples_per_second": 0.365,
      "eval_steps_per_second": 0.365,
      "step": 4
    },
    {
      "epoch": 0.01,
      "grad_norm": 3.3897781372070312,
      "learning_rate": 0.0002,
      "loss": 1.1807,
      "step": 5
    },
    {
      "epoch": 0.01,
      "eval_loss": 1.1253794431686401,
      "eval_runtime": 1918.161,
      "eval_samples_per_second": 0.365,
      "eval_steps_per_second": 0.365,
      "step": 5
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.9615206718444824,
      "learning_rate": 0.0002,
      "loss": 1.1417,
      "step": 6
    },
    {
      "epoch": 0.01,
      "eval_loss": 1.085837721824646,
      "eval_runtime": 1918.0897,
      "eval_samples_per_second": 0.365,
      "eval_steps_per_second": 0.365,
      "step": 6
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.8961493372917175,
      "learning_rate": 0.0002,
      "loss": 1.1129,
      "step": 7
    },
    {
      "epoch": 0.01,
      "eval_loss": 1.0493139028549194,
      "eval_runtime": 1918.1075,
      "eval_samples_per_second": 0.365,
      "eval_steps_per_second": 0.365,
      "step": 7
    }
  ],
  "logging_steps": 1,
  "max_steps": 10,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 1,
  "total_flos": 1226868850163712.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}