last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0005055931239335145,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 5.055931239335145e-05,
      "grad_norm": 2.8315072995610535e-05,
      "learning_rate": 2e-05,
      "loss": 46.0,
      "step": 1
    },
    {
      "epoch": 5.055931239335145e-05,
      "eval_loss": 11.5,
      "eval_runtime": 56.7046,
      "eval_samples_per_second": 146.867,
      "eval_steps_per_second": 73.433,
      "step": 1
    },
    {
      "epoch": 0.0001011186247867029,
      "grad_norm": 1.3820018466503825e-05,
      "learning_rate": 4e-05,
      "loss": 46.0,
      "step": 2
    },
    {
      "epoch": 0.00015167793718005436,
      "grad_norm": 1.7994578229263425e-05,
      "learning_rate": 6e-05,
      "loss": 46.0,
      "step": 3
    },
    {
      "epoch": 0.00015167793718005436,
      "eval_loss": 11.5,
      "eval_runtime": 56.516,
      "eval_samples_per_second": 147.357,
      "eval_steps_per_second": 73.678,
      "step": 3
    },
    {
      "epoch": 0.0002022372495734058,
      "grad_norm": 5.439726373879239e-05,
      "learning_rate": 8e-05,
      "loss": 46.0,
      "step": 4
    },
    {
      "epoch": 0.00025279656196675726,
      "grad_norm": 9.747568583406974e-06,
      "learning_rate": 0.0001,
      "loss": 46.0,
      "step": 5
    },
    {
      "epoch": 0.00030335587436010873,
      "grad_norm": 2.0666304408223368e-05,
      "learning_rate": 0.00012,
      "loss": 46.0,
      "step": 6
    },
    {
      "epoch": 0.00030335587436010873,
      "eval_loss": 11.5,
      "eval_runtime": 56.6901,
      "eval_samples_per_second": 146.904,
      "eval_steps_per_second": 73.452,
      "step": 6
    },
    {
      "epoch": 0.00035391518675346014,
      "grad_norm": 3.1393516110256314e-05,
      "learning_rate": 0.00014,
      "loss": 46.0,
      "step": 7
    },
    {
      "epoch": 0.0004044744991468116,
      "grad_norm": 1.6754234820837155e-05,
      "learning_rate": 0.00016,
      "loss": 46.0,
      "step": 8
    },
    {
      "epoch": 0.00045503381154016307,
      "grad_norm": 2.886553920689039e-05,
      "learning_rate": 0.00018,
      "loss": 46.0,
      "step": 9
    },
    {
      "epoch": 0.00045503381154016307,
      "eval_loss": 11.5,
      "eval_runtime": 56.8286,
      "eval_samples_per_second": 146.546,
      "eval_steps_per_second": 73.273,
      "step": 9
    },
    {
      "epoch": 0.0005055931239335145,
      "grad_norm": 5.010078166378662e-05,
      "learning_rate": 0.0002,
      "loss": 46.0,
      "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": 368573251584.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}