Question Answering
Transformers
Safetensors
Chinese
English
llama
text-generation
custom_code
text-generation-inference
Instructions to use FlagAlpha/Atom-7B-Chat with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use FlagAlpha/Atom-7B-Chat with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("question-answering", model="FlagAlpha/Atom-7B-Chat", trust_remote_code=True)# Load model directly from transformers import AutoTokenizer, AutoModelForCausalLM tokenizer = AutoTokenizer.from_pretrained("FlagAlpha/Atom-7B-Chat", trust_remote_code=True) model = AutoModelForCausalLM.from_pretrained("FlagAlpha/Atom-7B-Chat", trust_remote_code=True) - Notebooks
- Google Colab
- Kaggle
想要将下载的.safetensors模型文件导入到本地的ollama
#5
by hubblebubblepig - opened
想要将下载的.safetensors模型文件导入到本地的ollama,
按照ollama文档,其中一个步骤是需要对模型进行convert操作,报错:"NotImplementedError: Unknown rope scaling type: dynamic",
该如何处理?
更换使用convert-hf-to-gguf报错"Architecture "LlamaForCausalLM" not supported!"
在llama.cpp的issues中查到解决方法:
在convert-hf-to-gguf.py文件中
找到if arch == "MixtralForCausalLM" 添加 or arch == "LlamaForCausalLM"
想要将下载的.safetensors模型文件导入到本地的ollama,
按照ollama文档,其中一个步骤是需要对模型进行convert操作,报错:"NotImplementedError: Unknown rope scaling type: dynamic",
该如何处理?
更换使用convert-hf-to-gguf报错"Architecture "LlamaForCausalLM" not supported!"
我今天发现朋友的这个留言,我导入的时候没有提示这个问题。