IDEA-CCNL

/

Ziya-LLaMA-7B-Reward

like 65

1. reward.tolist()和reward.item()会报错，没有tolist()和item()方法
2. 不管有没有按照https://huggingface.co/IDEA-CCNL/Ziya-LLaMA-13B-v1#-%E4%BD%BF%E7%94%A8-usage- 中方法和原始llama模型合并后的模型，还是直接读取模型，都会出现下面信息

Some weights of the model checkpoint at data/Ziya-LLaMA-7B-Reward were not used when initializing LlamaForSequenceClassification: ['value_head.bias', 'value_head.weight']

• This IS expected if you are initializing LlamaForSequenceClassification from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model).
• This IS NOT expected if you are initializing LlamaForSequenceClassification from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model).
  Some weights of LlamaForSequenceClassification were not initialized from the model checkpoint at data/Ziya-LLaMA-7B-Reward and are newly initialized: ['score.weight']
  You should probably TRAIN this model on a down-stream task to be able to use it for predictions and inference.

3. 没有合并的原始模型用示例数据测试结果如下：
   
   合并后的模型数据测试结果如下：