evaluation_bloomz/evaluation_l2/xnli/de/justified_in_saying/results.json

{
  "dataset_name": "xnli",
  "dataset_config_name": "de",
  "template_name": "justified in saying",
  "evaluation": {
    "accuracy": 0.5248995983935743
  },
  "arguments": "Namespace(config_name=None, dataset_config_name='de', dataset_name='xnli', 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='en', template_name='justified in saying', tokenizer_name=None, use_slow_tokenizer=False)"
}