last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 8.432396423820676e-05,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 8.432396423820677e-06,
      "grad_norm": 0.43550553917884827,
      "learning_rate": 2e-05,
      "loss": 1.3887,
      "step": 1
    },
    {
      "epoch": 8.432396423820677e-06,
      "eval_loss": 1.1402034759521484,
      "eval_runtime": 2512.1721,
      "eval_samples_per_second": 19.876,
      "eval_steps_per_second": 9.938,
      "step": 1
    },
    {
      "epoch": 1.6864792847641354e-05,
      "grad_norm": 0.2166118025779724,
      "learning_rate": 4e-05,
      "loss": 1.0644,
      "step": 2
    },
    {
      "epoch": 2.529718927146203e-05,
      "grad_norm": 0.25337040424346924,
      "learning_rate": 6e-05,
      "loss": 0.7921,
      "step": 3
    },
    {
      "epoch": 2.529718927146203e-05,
      "eval_loss": 1.1398595571517944,
      "eval_runtime": 6028.2231,
      "eval_samples_per_second": 8.283,
      "eval_steps_per_second": 4.142,
      "step": 3
    },
    {
      "epoch": 3.372958569528271e-05,
      "grad_norm": 0.43934348225593567,
      "learning_rate": 8e-05,
      "loss": 1.2741,
      "step": 4
    },
    {
      "epoch": 4.216198211910338e-05,
      "grad_norm": 0.25466489791870117,
      "learning_rate": 0.0001,
      "loss": 0.8955,
      "step": 5
    },
    {
      "epoch": 5.059437854292406e-05,
      "grad_norm": 0.41991177201271057,
      "learning_rate": 0.00012,
      "loss": 1.2128,
      "step": 6
    },
    {
      "epoch": 5.059437854292406e-05,
      "eval_loss": 1.1368786096572876,
      "eval_runtime": 6918.8083,
      "eval_samples_per_second": 7.217,
      "eval_steps_per_second": 3.609,
      "step": 6
    },
    {
      "epoch": 5.9026774966744735e-05,
      "grad_norm": 0.38990527391433716,
      "learning_rate": 0.00014,
      "loss": 1.2169,
      "step": 7
    },
    {
      "epoch": 6.745917139056541e-05,
      "grad_norm": 0.304204523563385,
      "learning_rate": 0.00016,
      "loss": 1.0724,
      "step": 8
    },
    {
      "epoch": 7.58915678143861e-05,
      "grad_norm": 0.4411507844924927,
      "learning_rate": 0.00018,
      "loss": 1.2062,
      "step": 9
    },
    {
      "epoch": 7.58915678143861e-05,
      "eval_loss": 1.1268764734268188,
      "eval_runtime": 6915.961,
      "eval_samples_per_second": 7.22,
      "eval_steps_per_second": 3.61,
      "step": 9
    },
    {
      "epoch": 8.432396423820676e-05,
      "grad_norm": 0.3528674244880676,
      "learning_rate": 0.0002,
      "loss": 0.9766,
      "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": 1230072997478400.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}