MOSS
MOSS copied to clipboard
An open-source tool-augmented conversational language model from Fudan University
python3: /project/lib/Analysis/Utility.cpp:136: bool mlir::supportMMA(mlir::Value, int): Assertion `(version == 1 || version == 2) && "Unexpected MMA layout version found"' failed.
According to the exception message, it seems this issue possibly is caused by the missing implementation of Rust-based tokenization, while calling `tokenize()` method to tokenize batch of sequences and access...
在ModelScope上发布的moss-moon-003-sft-plugin模型,使用提供的Jupyter测试样例部署失败,在自己服务器和魔塔提供的实例上都不可以,错误是AttributeError: 'MossTokenizer' object has no attribute 'encoder',GPU用的A40不知道有没有影响 `AttributeError Traceback (most recent call last) File /opt/conda/lib/python3.10/site-packages/modelscope/utils/registry.py:212, in build_from_cfg(cfg, registry, group_key, default_args) 211 else: --> 212 return obj_cls(**args) 213 except Exception as...
550W
当下v100和rtx有点难搞,我能不能用别的?
Fix #159
This pull request fixes the TypeError when doing inference with [moss-moon-003-sft-int4](https://huggingface.co/fnlp/moss-moon-003-sft-int4), specifically,`TypeError: ' Union[Tuple[torch.Tensor], Optional[Tuple[torch.Tensor, Tuple[torch.FloatTensor │ │ 268 │ │ residual = hidden_states │ │ 269 │ │ hidden_states...
您好 readme中提到: `moss-moon-003-sft所使用的多轮对话数据,基于MOSS-002内测阶段采集的约10万用户输入数据和gpt-3.5-turbo构造而成` 请问是将user prompt输入到chatgpt中一轮一轮来增量构造的吗? 那么是否会存在用户在第二轮提的内容在gpt第一轮中没有出现过,比如下面的示例: ``` user: 给我写一个快排 gpt: code..... user: 你的代码里面的quicksort函数是什么意思 gpt:对不起我之前的回答里面并没有提到quicksort这个函数 ``` 这种情况是不是上下文的语义不太统一,开源的数据里面考虑过这种问题吗?
求一個moss邀請碼,郵件:[email protected]