run-4/checkpoint-1500/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 4.716981132075472,
  "eval_steps": 500,
  "global_step": 1500,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5735483870967742,
      "eval_loss": 0.20021426677703857,
      "eval_runtime": 5.4374,
      "eval_samples_per_second": 570.126,
      "eval_steps_per_second": 11.954,
      "step": 318
    },
    {
      "epoch": 1.5723270440251573,
      "grad_norm": 0.5314117670059204,
      "learning_rate": 1.4758909853249476e-05,
      "loss": 0.3171,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.81,
      "eval_loss": 0.10166090726852417,
      "eval_runtime": 5.5055,
      "eval_samples_per_second": 563.069,
      "eval_steps_per_second": 11.806,
      "step": 636
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8670967741935484,
      "eval_loss": 0.07133448123931885,
      "eval_runtime": 5.5034,
      "eval_samples_per_second": 563.284,
      "eval_steps_per_second": 11.811,
      "step": 954
    },
    {
      "epoch": 3.1446540880503147,
      "grad_norm": 0.45540565252304077,
      "learning_rate": 9.517819706498952e-06,
      "loss": 0.1173,
      "step": 1000
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8896774193548387,
      "eval_loss": 0.05815433710813522,
      "eval_runtime": 5.5711,
      "eval_samples_per_second": 556.446,
      "eval_steps_per_second": 11.667,
      "step": 1272
    },
    {
      "epoch": 4.716981132075472,
      "grad_norm": 0.3324480354785919,
      "learning_rate": 4.276729559748428e-06,
      "loss": 0.0807,
      "step": 1500
    }
  ],
  "logging_steps": 500,
  "max_steps": 1908,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 6,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 391368939443328.0,
  "train_batch_size": 48,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.760358408342671,
    "num_train_epochs": 6,
    "temperature": 14
  }
}