last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.004565688848304988,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0004565688848304988,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 6.1476,
      "step": 1
    },
    {
      "epoch": 0.0004565688848304988,
      "eval_loss": 8.7471342086792,
      "eval_runtime": 60.3923,
      "eval_samples_per_second": 7.65,
      "eval_steps_per_second": 7.65,
      "step": 1
    },
    {
      "epoch": 0.0009131377696609977,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 8.1898,
      "step": 2
    },
    {
      "epoch": 0.0013697066544914964,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 6.0634,
      "step": 3
    },
    {
      "epoch": 0.0013697066544914964,
      "eval_loss": 8.7471342086792,
      "eval_runtime": 60.3302,
      "eval_samples_per_second": 7.658,
      "eval_steps_per_second": 7.658,
      "step": 3
    },
    {
      "epoch": 0.0018262755393219953,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 7.7222,
      "step": 4
    },
    {
      "epoch": 0.002282844424152494,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 7.0675,
      "step": 5
    },
    {
      "epoch": 0.002739413308982993,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 7.9791,
      "step": 6
    },
    {
      "epoch": 0.002739413308982993,
      "eval_loss": 8.7471342086792,
      "eval_runtime": 60.3363,
      "eval_samples_per_second": 7.657,
      "eval_steps_per_second": 7.657,
      "step": 6
    },
    {
      "epoch": 0.0031959821938134917,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 8.8075,
      "step": 7
    },
    {
      "epoch": 0.0036525510786439906,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 9.0902,
      "step": 8
    },
    {
      "epoch": 0.004109119963474489,
      "grad_norm": 34.917747497558594,
      "learning_rate": 2e-05,
      "loss": 6.32,
      "step": 9
    },
    {
      "epoch": 0.004109119963474489,
      "eval_loss": 8.7471342086792,
      "eval_runtime": 59.8996,
      "eval_samples_per_second": 7.713,
      "eval_steps_per_second": 7.713,
      "step": 9
    },
    {
      "epoch": 0.004565688848304988,
      "grad_norm": 68.01629638671875,
      "learning_rate": 4e-05,
      "loss": 7.7139,
      "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": 2010596581048320.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}