MiniCPM

MiniCPM copied to clipboard

### 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 /...

TianmengChen

[Bad Case]: error loading model architecture: unknown model architecture: 'minicpm3' time=2024-10-12T20:12:01.292+08:00 level=ERROR source=sched.go:456 msg="error loading llama server" error="llama runner process has terminated: this model is not supported by your version of Ollama. You may need to upgrade"

3

### Description / 描述 error loading model architecture: unknown model architecture: 'minicpm3' time=2024-10-12T20:12:01.292+08:00 level=ERROR source=sched.go:456 msg="error loading llama server" error="llama runner process has terminated: this model is not supported by...

Hoyxxx

[Feature Request]: 建议将finetune里的.sh文件中的--include localhost:1更改为--include localhost:0

1

### Feature request / 功能建议 大部分人在微调2b模型用的是单卡，建议默认参数为--include localhost:0

wsstudent

[Feature Request]: 如果使用CPU的话，要求的CPU核数和内存大概是多大，运行4B的模型。

1

### Feature request / 功能建议 如果使用CPU的话，要求的CPU核数和内存大概是多大，运行4B的模型。

zaojiahua

### Feature request / 功能建议 后续有开源Minicpm量化模型相关代码的计划吗。

eureka336

Any plan for MiniCPMForTokenClassification?

Evilran

### 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 /...

GladiousZhang

Thank you for releasing the excellent model and work. The paper appears to state that the 8B model was pre-trained natively with a 32K sequence length. I would like to...

wavy-jung

请问你们开源的代码中，在哪里能找到sparse attention的torch实现吗

andy-upup

[Bad Case]: 使用minicpm4-0.5B同时启用sparse attention时报错

3

### Description / 描述 你好！ 我注意到minicpm4-0.5B中描述无法支持sparse attention，但是我看modeling中有就修改为可用，但报错： ``` topk_idx[topk_idx >= q_idx[None, :, None]] = -1 RuntimeError: The size of tensor a (355) must match the size of tensor b (711)...

junyou2001