trainer_state.json

{
  "best_global_step": 20,
  "best_metric": 0.3098881333544819,
  "best_model_checkpoint": "/content/gemma_lora_imb/checkpoint-20",
  "epoch": 1.253968253968254,
  "eval_steps": 20,
  "global_step": 20,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.253968253968254,
      "grad_norm": 164.26077270507812,
      "learning_rate": 5.9375e-06,
      "loss": 9.5154,
      "step": 20
    },
    {
      "epoch": 1.253968253968254,
      "eval_f1_macro": 0.3098881333544819,
      "eval_loss": 1.8632150888442993,
      "eval_runtime": 11.091,
      "eval_samples_per_second": 45.082,
      "eval_steps_per_second": 5.68,
      "step": 20
    }
  ],
  "logging_steps": 20,
  "max_steps": 32,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 20,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 329551733809152.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}