checkpoint-2000/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.5288207297726071,
  "eval_steps": 1000,
  "global_step": 2000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.13220518244315177,
      "grad_norm": 0.8546391725540161,
      "learning_rate": 8.816009873931059e-05,
      "loss": 5.1118,
      "step": 500
    },
    {
      "epoch": 0.26441036488630354,
      "grad_norm": 0.8593688607215881,
      "learning_rate": 9.59831475011252e-05,
      "loss": 3.406,
      "step": 1000
    },
    {
      "epoch": 0.26441036488630354,
      "eval_accuracy": 0.5035306174465283,
      "eval_loss": 3.23445987701416,
      "eval_runtime": 74.0522,
      "eval_samples_per_second": 24.847,
      "eval_steps_per_second": 0.527,
      "step": 1000
    },
    {
      "epoch": 0.3966155473294553,
      "grad_norm": 0.9617258906364441,
      "learning_rate": 9.134314230431938e-05,
      "loss": 3.0005,
      "step": 1500
    },
    {
      "epoch": 0.5288207297726071,
      "grad_norm": 0.8953185677528381,
      "learning_rate": 8.670313710751356e-05,
      "loss": 2.8119,
      "step": 2000
    },
    {
      "epoch": 0.5288207297726071,
      "eval_accuracy": 0.5365118094348038,
      "eval_loss": 2.821384906768799,
      "eval_runtime": 72.8065,
      "eval_samples_per_second": 25.272,
      "eval_steps_per_second": 0.536,
      "step": 2000
    }
  ],
  "logging_steps": 500,
  "max_steps": 11346,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 1000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 2.5630334779392e+16,
  "train_batch_size": 24,
  "trial_name": null,
  "trial_params": null
}