openchat
openchat copied to clipboard
OpenChat: Advancing Open-source Language Models with Imperfect Data
huggingface -> Hugging Face
Great Works! It seems that [openchat.train.json](https://huggingface.co/datasets/openchat/openchat_sharegpt4_dataset/blob/main/openchat.train.json) does not utilize a prompt template like what [alpaca-lora](https://github.com/tloen/alpaca-lora/tree/main/templates) does. Do you make experiments about using a prompt template? Will that be better or...
And can you describe the details of a conditioning strategy and weighted loss? Thanks!
I am getting different errors. For example AssertionError: pydantic.dataclasses.dataclass only supports init=False I have to downgrade pydantic version. It would be great if you can add all packages versions
Trying to install it to NVidia's pytorch contaner. I'm getting this while running. Same issue while trying to install it to Lambda GPU cloud on H100 instance. (all default) ```...
Great job! I have a few questions: 1. I'm using the following script to test OpenChat, but even with the correct prompt template, the output is not very accurate. How...
would be great to setup a gradio demo for this on huggingface, similar to https://huggingface.co/spaces/mosaicml/mpt-30b-chat, this is the guide: https://huggingface.co/docs/hub/spaces-sdks-gradio
when i click on this link: https://huggingface.co/datasets/openchat/openchat_sharegpt4_dataset It says no able to see the data:
Hi, may you provide the detailed hyper-paramters when you training llama-13b? For example, how many and what kind of GPUs you use, what are the gradient accumulation steps and batch...