last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0014849034812737172,
  "eval_steps": 3,
  "global_step": 9,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00016498927569707968,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 9.7696,
      "step": 1
    },
    {
      "epoch": 0.00016498927569707968,
      "eval_loss": 10.010821342468262,
      "eval_runtime": 268.3535,
      "eval_samples_per_second": 4.759,
      "eval_steps_per_second": 4.759,
      "step": 1
    },
    {
      "epoch": 0.00032997855139415937,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 9.9709,
      "step": 2
    },
    {
      "epoch": 0.000494967827091239,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 10.2079,
      "step": 3
    },
    {
      "epoch": 0.000494967827091239,
      "eval_loss": 10.010821342468262,
      "eval_runtime": 268.7546,
      "eval_samples_per_second": 4.752,
      "eval_steps_per_second": 4.752,
      "step": 3
    },
    {
      "epoch": 0.0006599571027883187,
      "grad_norm": 21.604045867919922,
      "learning_rate": 2e-05,
      "loss": 9.7673,
      "step": 4
    },
    {
      "epoch": 0.0008249463784853984,
      "grad_norm": 23.223350524902344,
      "learning_rate": 4e-05,
      "loss": 10.0461,
      "step": 5
    },
    {
      "epoch": 0.000989935654182478,
      "grad_norm": 21.16267204284668,
      "learning_rate": 6e-05,
      "loss": 9.6332,
      "step": 6
    },
    {
      "epoch": 0.000989935654182478,
      "eval_loss": 8.121424674987793,
      "eval_runtime": 265.9905,
      "eval_samples_per_second": 4.801,
      "eval_steps_per_second": 4.801,
      "step": 6
    },
    {
      "epoch": 0.0011549249298795578,
      "grad_norm": 21.66244888305664,
      "learning_rate": 8e-05,
      "loss": 8.14,
      "step": 7
    },
    {
      "epoch": 0.0013199142055766375,
      "grad_norm": 22.188928604125977,
      "learning_rate": 0.0001,
      "loss": 6.2351,
      "step": 8
    },
    {
      "epoch": 0.0014849034812737172,
      "grad_norm": 16.275470733642578,
      "learning_rate": 0.00012,
      "loss": 4.7693,
      "step": 9
    },
    {
      "epoch": 0.0014849034812737172,
      "eval_loss": 2.870197057723999,
      "eval_runtime": 266.0215,
      "eval_samples_per_second": 4.8,
      "eval_steps_per_second": 4.8,
      "step": 9
    }
  ],
  "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": false
      },
      "attributes": {}
    }
  },
  "total_flos": 6676985552044032.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}