William

Results 9 issues of William

### info break相关 关于info b中的信息: disp一列中`keep`和`dis`和`del`, 对应断点生效后是 保留 还是 disable 还是 delete enb一列中`y`和`n`,对应该断点现在disable还是enable 例如: disable 1 会让断点失效(disp=dis, enb=n),之后enable once 1 会让(disp=dis, enb=y), 然后该断点生效一次之后(disp=dis, enb=n) 又例如:tbreak main后(disp=del, enb=y) ### break if...

exclude prefix start with '.', so that no common structure could be matched.

### Is there an existing issue for this? - [X] I have searched the existing issues ### Description of the Bug TP linear async_chunk=4 mismatch async_chunk=1 result when sequence length...

bug

Can you please upload your pruning mask related to [Sheared-LLaMA-1.3B-Pruned](https://huggingface.co/princeton-nlp/Sheared-LLaMA-1.3B-Pruned) and [Sheared-LLaMA-2.7B-Pruned](https://huggingface.co/princeton-nlp/Sheared-LLaMA-2.7B-Pruned)? This would help researchers conduct further analysis on weight importance, sparsity, etc.

I created a folder called "minicpmv" in the examples folder of llama.cpp. More detail can be seen in `llama.cpp/examples/minicpmv/README.md`. The code is based on `examples/llava` but the vision part is...

enhancement
demo
examples
python
Review Complexity : High

In the `README.md`, you mentioned that > The data preparation code for self-distillation can be found in [data_generation folder](https://github.com/FasterDecoding/Medusa/blob/main/data_generation) of the current repo. In that folder, it says > `python...