last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.004730928444707274,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0004730928444707274,
      "grad_norm": NaN,
      "learning_rate": 2e-05,
      "loss": 0.0,
      "step": 1
    },
    {
      "epoch": 0.0004730928444707274,
      "eval_loss": NaN,
      "eval_runtime": 20.4076,
      "eval_samples_per_second": 43.611,
      "eval_steps_per_second": 21.806,
      "step": 1
    },
    {
      "epoch": 0.0009461856889414548,
      "grad_norm": NaN,
      "learning_rate": 4e-05,
      "loss": 2.5373,
      "step": 2
    },
    {
      "epoch": 0.0014192785334121822,
      "grad_norm": NaN,
      "learning_rate": 6e-05,
      "loss": 1.4898,
      "step": 3
    },
    {
      "epoch": 0.0014192785334121822,
      "eval_loss": NaN,
      "eval_runtime": 15.5399,
      "eval_samples_per_second": 57.272,
      "eval_steps_per_second": 28.636,
      "step": 3
    },
    {
      "epoch": 0.0018923713778829095,
      "grad_norm": NaN,
      "learning_rate": 8e-05,
      "loss": 0.0,
      "step": 4
    },
    {
      "epoch": 0.002365464222353637,
      "grad_norm": NaN,
      "learning_rate": 0.0001,
      "loss": 0.0,
      "step": 5
    },
    {
      "epoch": 0.0028385570668243644,
      "grad_norm": NaN,
      "learning_rate": 0.00012,
      "loss": 0.0,
      "step": 6
    },
    {
      "epoch": 0.0028385570668243644,
      "eval_loss": NaN,
      "eval_runtime": 15.6748,
      "eval_samples_per_second": 56.779,
      "eval_steps_per_second": 28.39,
      "step": 6
    },
    {
      "epoch": 0.0033116499112950917,
      "grad_norm": NaN,
      "learning_rate": 0.00014,
      "loss": 0.0,
      "step": 7
    },
    {
      "epoch": 0.003784742755765819,
      "grad_norm": NaN,
      "learning_rate": 0.00016,
      "loss": 1.5359,
      "step": 8
    },
    {
      "epoch": 0.004257835600236547,
      "grad_norm": NaN,
      "learning_rate": 0.00018,
      "loss": 1.4217,
      "step": 9
    },
    {
      "epoch": 0.004257835600236547,
      "eval_loss": NaN,
      "eval_runtime": 15.0413,
      "eval_samples_per_second": 59.17,
      "eval_steps_per_second": 29.585,
      "step": 9
    },
    {
      "epoch": 0.004730928444707274,
      "grad_norm": NaN,
      "learning_rate": 0.0002,
      "loss": 6.0998,
      "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": 340116300103680.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}