last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.00047441706003747893,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 4.74417060037479e-05,
      "grad_norm": NaN,
      "learning_rate": 2e-05,
      "loss": 0.0,
      "step": 1
    },
    {
      "epoch": 4.74417060037479e-05,
      "eval_loss": NaN,
      "eval_runtime": 366.7755,
      "eval_samples_per_second": 24.2,
      "eval_steps_per_second": 12.1,
      "step": 1
    },
    {
      "epoch": 9.48834120074958e-05,
      "grad_norm": NaN,
      "learning_rate": 4e-05,
      "loss": 0.0,
      "step": 2
    },
    {
      "epoch": 0.00014232511801124368,
      "grad_norm": NaN,
      "learning_rate": 6e-05,
      "loss": 0.0,
      "step": 3
    },
    {
      "epoch": 0.00014232511801124368,
      "eval_loss": NaN,
      "eval_runtime": 366.1777,
      "eval_samples_per_second": 24.24,
      "eval_steps_per_second": 12.12,
      "step": 3
    },
    {
      "epoch": 0.0001897668240149916,
      "grad_norm": NaN,
      "learning_rate": 8e-05,
      "loss": 0.0,
      "step": 4
    },
    {
      "epoch": 0.00023720853001873946,
      "grad_norm": NaN,
      "learning_rate": 0.0001,
      "loss": 0.0,
      "step": 5
    },
    {
      "epoch": 0.00028465023602248737,
      "grad_norm": NaN,
      "learning_rate": 0.00012,
      "loss": 0.0,
      "step": 6
    },
    {
      "epoch": 0.00028465023602248737,
      "eval_loss": NaN,
      "eval_runtime": 366.3866,
      "eval_samples_per_second": 24.226,
      "eval_steps_per_second": 12.113,
      "step": 6
    },
    {
      "epoch": 0.00033209194202623525,
      "grad_norm": NaN,
      "learning_rate": 0.00014,
      "loss": 0.0,
      "step": 7
    },
    {
      "epoch": 0.0003795336480299832,
      "grad_norm": NaN,
      "learning_rate": 0.00016,
      "loss": 0.0,
      "step": 8
    },
    {
      "epoch": 0.00042697535403373105,
      "grad_norm": NaN,
      "learning_rate": 0.00018,
      "loss": 0.0,
      "step": 9
    },
    {
      "epoch": 0.00042697535403373105,
      "eval_loss": NaN,
      "eval_runtime": 366.4724,
      "eval_samples_per_second": 24.22,
      "eval_steps_per_second": 12.11,
      "step": 9
    },
    {
      "epoch": 0.00047441706003747893,
      "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": 557300091715584.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}