trainer_state.json

{
  "best_metric": 3.6337857246398926,
  "best_model_checkpoint": "./qwen2vl_medical_finetuned_from_step4/checkpoint-1",
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 3,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_loss": 3.6337857246398926,
      "eval_mean_token_accuracy": 0.33830709755420685,
      "eval_num_tokens": 2131.0,
      "eval_runtime": 4.7827,
      "eval_samples_per_second": 1.045,
      "eval_steps_per_second": 0.418,
      "step": 1
    },
    {
      "epoch": 2.0,
      "eval_loss": 3.6362528800964355,
      "eval_mean_token_accuracy": 0.339595764875412,
      "eval_num_tokens": 4262.0,
      "eval_runtime": 4.7327,
      "eval_samples_per_second": 1.056,
      "eval_steps_per_second": 0.423,
      "step": 2
    },
    {
      "epoch": 3.0,
      "eval_loss": 3.636836528778076,
      "eval_mean_token_accuracy": 0.339595764875412,
      "eval_num_tokens": 6393.0,
      "eval_runtime": 4.7295,
      "eval_samples_per_second": 1.057,
      "eval_steps_per_second": 0.423,
      "step": 3
    },
    {
      "epoch": 3.0,
      "mean_token_accuracy": 0.38623279001977706,
      "num_tokens": 6393.0,
      "step": 3,
      "total_flos": 77489850439680.0,
      "train_loss": 1.2701127529144287,
      "train_runtime": 37.7856,
      "train_samples_per_second": 0.397,
      "train_steps_per_second": 0.079
    }
  ],
  "logging_steps": 20,
  "max_steps": 3,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 77489850439680.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}