trainer_state.json

{
  "best_metric": 0.034021906554698944,
  "best_model_checkpoint": "models/HG/checkpoint-600",
  "epoch": 3.045685279187817,
  "eval_steps": 200,
  "global_step": 600,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.015228426395939,
      "grad_norm": 0.2426794022321701,
      "learning_rate": 6.349206349206349e-06,
      "loss": 0.0262,
      "step": 200
    },
    {
      "epoch": 1.015228426395939,
      "eval_loss": 0.037922926247119904,
      "eval_runtime": 1.4442,
      "eval_samples_per_second": 484.703,
      "eval_steps_per_second": 15.234,
      "step": 200
    },
    {
      "epoch": 2.030456852791878,
      "grad_norm": 0.3046327233314514,
      "learning_rate": 1.2698412698412699e-05,
      "loss": 0.0159,
      "step": 400
    },
    {
      "epoch": 2.030456852791878,
      "eval_loss": 0.03600747138261795,
      "eval_runtime": 1.4409,
      "eval_samples_per_second": 485.792,
      "eval_steps_per_second": 15.268,
      "step": 400
    },
    {
      "epoch": 3.045685279187817,
      "grad_norm": 0.08811034262180328,
      "learning_rate": 1.904761904761905e-05,
      "loss": 0.0082,
      "step": 600
    },
    {
      "epoch": 3.045685279187817,
      "eval_loss": 0.034021906554698944,
      "eval_runtime": 1.4347,
      "eval_samples_per_second": 487.892,
      "eval_steps_per_second": 15.334,
      "step": 600
    }
  ],
  "logging_steps": 200,
  "max_steps": 985,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 200,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}