trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0007727975270479134,
  "eval_steps": 500,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "learning_rate": 0.0001,
      "loss": 2.0357,
      "step": 1
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0002,
      "loss": 1.8357,
      "step": 2
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.000175,
      "loss": 1.6356,
      "step": 3
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.00015000000000000001,
      "loss": 1.3577,
      "step": 4
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.000125,
      "loss": 1.3596,
      "step": 5
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0001,
      "loss": 1.0625,
      "step": 6
    },
    {
      "epoch": 0.0,
      "learning_rate": 7.500000000000001e-05,
      "loss": 1.0769,
      "step": 7
    },
    {
      "epoch": 0.0,
      "learning_rate": 5e-05,
      "loss": 1.2269,
      "step": 8
    },
    {
      "epoch": 0.0,
      "learning_rate": 2.5e-05,
      "loss": 1.0216,
      "step": 9
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0,
      "loss": 0.9863,
      "step": 10
    },
    {
      "epoch": 0.0,
      "step": 10,
      "total_flos": 324056878080000.0,
      "train_loss": 1.359846395254135,
      "train_runtime": 36.0797,
      "train_samples_per_second": 1.109,
      "train_steps_per_second": 0.277
    }
  ],
  "logging_steps": 1,
  "max_steps": 10,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 500,
  "total_flos": 324056878080000.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}