evaluation_bloomz/evaluation_val/aqua_rat/raw/answer_quiz/results.json

{
  "dataset_name": "aqua_rat",
  "dataset_config_name": "raw",
  "template_name": "answer_quiz",
  "evaluation": {
    "accuracy": 0.2677165354330709
  },
  "arguments": "Namespace(config_name=None, dataset_config_name='raw', dataset_name='aqua_rat', debug=False, dtype='bfloat16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/xp3capmixnewcodelonglossseq_global_step498', output_dir='/gpfsscratch/rech/six/commun/commun/experiments/muennighoff/bloomckpt/176bt0/xp3capmixnewcodelonglossseq_global_step498/evaluation', pad_to_max_length=False, per_device_eval_batch_size=8, prefixlm=False, split='validation', target_max_length=256, template_config_name=None, template_name='answer_quiz', tokenizer_name=None, use_slow_tokenizer=False)"
}