MiniCPM

MiniCPM copied to clipboard

[Feature Request]:ollama怎么部署新版v2

2

### Feature request / 功能建议 ollama怎么部署新版v2

yuanjie-ai

您好， 请问如果想在你们模型的基础上继续在某个领域的数据(大概20B）上继续预训练，是不是可以在你们退火后的学习率1e-3的基础上使用WSD的学习策略？是将1e-3设置为最大学习率，然后运用你们的公式根据步数调整LR吗？

mynewstart

### Is there an existing issue ? / 是否已有相关的 issue ? - [X] I have searched, and there is no existing issue. / 我已经搜索过了，没有相关的 issue。 ### Describe the bug /...

sankexin

### Is there an existing issue ? / 是否已有相关的 issue ? - [X] I have searched, and there is no existing issue. / 我已经搜索过了，没有相关的 issue。 ### Describe the bug /...

tzx666

```python def __init__( self, data_path, tokenizer, model_max_length=4096, user_tokens=[1786, 4194, 95388], assistant_tokens=[1786, 10850, 95388], ): ``` We try to do sft on `MiniCPM-1B-sft-bp16`, but there is an error, because the token...

PUITAR

用Linux transformers载入推理的，测试了常规对话，在跑的时候没有看到支持 Function Call 等功能

zRzRzRzRzRzRzR

### Is there an existing issue ? / 是否已有相关的 issue ? - [X] I have searched, and there is no existing issue. / 我已经搜索过了，没有相关的 issue。 ### Describe the bug /...

lonelydancer

### Feature request / 功能建议 Seems only the text part is supported on llama.cpp, but the multimodal part is not. It would be really helpful if it could be supported,...

yyyoungman

想问一下 Ultrachat/sharegpt 放入预训练的数据拼接方式是怎样的呢？

Damilytutu

### Feature request / 功能建议 Thanks for your great work. Can you provide more details on the proxy model? I have several questions. 1. Depth scale on the attention/mlp sub-block,...

yzlnew