白石

图模型章节的公式 14.5 中 x'A 和xA 有什么不同的喔？ x' 是 x 的补集吗？ 还是仅仅只是为了将 x‘ 和 x 区分开，每什么其他意义？

公式14.11 中 j 和 k 分别表示的是什么意思喔？ j是指转移特征函数的 种类吗？ k 是指状态特征函数的种类 吗？

you can sea example of multi gpu for inference from https://huggingface.co/OpenGVLab/InternVL-Chat-V1-5. And similar solution from https://github.com/OpenGVLab/InternVL/issues/96, and A part of the code requires modification. ``` # modeling_internval_chat.py line 353 input_embeds[selected]...

you can sea example of multi gpu for inference from https://huggingface.co/OpenGVLab/InternVL-Chat-V1-5. ``` # Otherwise, you need to set device_map='auto' to use multiple GPUs for inference. model = AutoModel.from_pretrained( path, torch_dtype=torch.bfloat16,...

In order to solve the problem of "FlashAttention only supports Ampere GPUs or newer". In this way, it can be solved.  change the content of config.json in [InternVL-Chat-V1-2-Plus](https://huggingface.co/OpenGVLab/InternVL-Chat-V1-2-Plus): 1....

Copying the code of `def stream_chat(self, ....)` to `modeling_internvl_chat.py` from `modeling_internlm2.py` [https://huggingface.co/OpenGVLab/InternVL-Chat-V1-5/blob/main/modeling_internlm2.py](url) and make very small changes , I implement it and verify that is usefull.