abetlen / abetlen/llama-cpp-python

A different value of n_batch gives different logits (llama.cpp_hf)

Đang mở
#1,018 0 bình luận 0 reaction 0 người được giao Xem trên GitHub
bug
Ngôn ngữ chính
Python
Star
10.6k
Fork
1.4k
Chỉ số merge pull request
Chỉ số pull request đang chờ

Mô tả

Hello,

I was told that n_batch doesn't alter the logits but after doing some tests I can see that it's not working as intented (or is it?).

"*When was born Tony Yoka? - Tony Yoka was born the 28 April 199...*"
The next token is supposed to be "2"
![N8IiAaQ](https://github.com/oobabooga/text-generation-webui/assets/110173477/b854ad25-c5a6-4a3c-a404-832ee14a43c3)

You can see that a low n_batch value objectively leads to a deterioration in the quality of the results (The only viable token is less likely to be chosen).

### Is there an existing issue for this?

- [X] I have searched the existing issues

### Reproduction

Change the n_batch value and compare the logits.

### Screenshot

_No response_

### Logs

```shell
No logs.
```

### System Info

```shell
Windows 10.
RTX 3060.
```

Hướng dẫn đóng góp

Mở hướng dẫn đóng góp

Đánh giá

Issue này chưa được đánh giá.

Nhận issue mới trong hộp thư của bạn

Bản tóm tắt ngắn những issue GitHub phù hợp với người mới.