FastChat
FastChat copied to clipboard
lm-sys
An open platform for training, serving, and evaluating large language models. Release repo for Vicuna and Chatbot Arena.
Not an issue exactly. But couldn't figure. In training scrip, I believe it replaces all the multi-head attn with flash attn but not sure about what's happening in inference. Any...
after installing everything i have garbage text on any question that i send to chat. 
can i use this code to conduct self-supervised pre-training based on llama-7b? (Not the dialog format given in the example, My dataset is pieces of text) if i can, which...
What are all the languages present in the ShareGPT 70,000 conversation dataset which was used to fine-tune FastChat-T5? The ReadMe file points to [`data_cleaning.md`](https://github.com/lm-sys/FastChat/blob/main/docs/commands/data_cleaning.md) which was used to get data...
RuntimeError: Expected all tensors to be on the same device, but found at least two devices, cuda:1 and cuda:0!
Hi :wave: I was wondering if there is any ongoing incentives for training a new Vicuna model based on the fully open source [OpenLLaMA](https://github.com/openlm-research/open_llama)? This would ultimately remove the requirement...
fastchat==0.2.9 transformers is also the newest (commit a2789addd) vicuna-13b-delta-v1.1 
Does the Fine-tuning Vicuna-7B with Local GPUs is full parameter fine-tuning? Does it use the rola?
Finetune 13b model on V100 with bs1 in int8 got OOM? I already enabled Deepspeed
Hello LMSYS Team, I am reaching out to inquire about the functionality of the Vicuna model. I would like to know if there is an available API for Vicuna that...
