晨星 comments

Results 6 comments of


                                            晨星

<titleExplicitly passing a `revision`>

我也出现了！ ``` Explicitly passing a `revision` is encouraged when loading a model with custom code to ensure no malicious code has been contributed in a newer revision. Explicitly passing a...

<titleExplicitly passing a `revision`>

我自己试了一下加这个加粗部分能解决一个报错( tokenizer = AutoTokenizer.from_pretrained(******, trust_remote_code=True **, revision="main"**)

<titleExplicitly passing a `revision`>

解决方法(自己使用成功了)：加" , revision="main" ": `tokenizer = AutoTokenizer.from_pretrained(******, trust_remote_code=True , revision="main")` 重装torch老版本: `pip install torch==1.12.0`

[llama-3.1 70B]Open Interpreter's Preps did not complete after setting the model

same too

[llama-3.1 70B]Open Interpreter's Preps did not complete after setting the model

It seems doesn't supported llama3.1

[llama-3.1 70B]Open Interpreter's Preps did not complete after setting the model

Maybe you can run `llama run llama3` and use `interpreter --model ollama/llama3` to use