last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.023952095808383235,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0023952095808383233,
      "grad_norm": NaN,
      "learning_rate": 2e-05,
      "loss": 0.0,
      "step": 1
    },
    {
      "epoch": 0.0023952095808383233,
      "eval_loss": NaN,
      "eval_runtime": 5.1301,
      "eval_samples_per_second": 34.307,
      "eval_steps_per_second": 17.154,
      "step": 1
    },
    {
      "epoch": 0.004790419161676647,
      "grad_norm": NaN,
      "learning_rate": 4e-05,
      "loss": 0.0,
      "step": 2
    },
    {
      "epoch": 0.00718562874251497,
      "grad_norm": NaN,
      "learning_rate": 6e-05,
      "loss": 0.0,
      "step": 3
    },
    {
      "epoch": 0.00718562874251497,
      "eval_loss": NaN,
      "eval_runtime": 4.1072,
      "eval_samples_per_second": 42.851,
      "eval_steps_per_second": 21.426,
      "step": 3
    },
    {
      "epoch": 0.009580838323353293,
      "grad_norm": NaN,
      "learning_rate": 8e-05,
      "loss": 0.0,
      "step": 4
    },
    {
      "epoch": 0.011976047904191617,
      "grad_norm": NaN,
      "learning_rate": 0.0001,
      "loss": 0.0,
      "step": 5
    },
    {
      "epoch": 0.01437125748502994,
      "grad_norm": NaN,
      "learning_rate": 0.00012,
      "loss": 0.0,
      "step": 6
    },
    {
      "epoch": 0.01437125748502994,
      "eval_loss": NaN,
      "eval_runtime": 4.1238,
      "eval_samples_per_second": 42.679,
      "eval_steps_per_second": 21.339,
      "step": 6
    },
    {
      "epoch": 0.016766467065868262,
      "grad_norm": NaN,
      "learning_rate": 0.00014,
      "loss": 0.0,
      "step": 7
    },
    {
      "epoch": 0.019161676646706587,
      "grad_norm": NaN,
      "learning_rate": 0.00016,
      "loss": 0.0,
      "step": 8
    },
    {
      "epoch": 0.02155688622754491,
      "grad_norm": NaN,
      "learning_rate": 0.00018,
      "loss": 0.0,
      "step": 9
    },
    {
      "epoch": 0.02155688622754491,
      "eval_loss": NaN,
      "eval_runtime": 4.136,
      "eval_samples_per_second": 42.553,
      "eval_steps_per_second": 21.276,
      "step": 9
    },
    {
      "epoch": 0.023952095808383235,
      "grad_norm": NaN,
      "learning_rate": 0.0002,
      "loss": 0.0,
      "step": 10
    }
  ],
  "logging_steps": 1,
  "max_steps": 10,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 3,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 28035804561408.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}