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.
Hi Team, Can someone please provide the command line instruction to execute train_lora.py file as the readme only contains the command for train_mem.py Thank you
**!!! It's not an official WeChat Group, if the lm-sys find that it's not proper to post it, contact me to delete this.** A group for people who are interested...
I'm using CPU to run train_lora.py. The problem is each worker process needs to load an independent model copy into memory. You know that LoRA won't change the model weight,...
can anyone tell me that how can i finetune the model use just single gpu???
Could you please provide information about the fine-tuning data used for the vicuna-13b-delta-v1.1 weight: 1. Is it fine-tuned on multi-turn dialogues or single-turn dialogues? 2. Does it include data with...
RuntimeError: The size of tensor a (65537024) must match the size of tensor b (262148096) at non-singleton dimension 0 site-packages/torch/optim/adamw.py:273 in │ │ _single_tensor_adamw │ │ │ │ 270 │...
When going through the steps for Web UI, I get the following output in the console: When I open in the browser I get this. Any quick solutions for this?...
I found that when fine-tuning the 7B model, there were no error messages or interruptions, but not all parameters were saved completely. Was a multi-GPU V100 used torchrun --nproc_per_node=8 --master_port=20001...
Maybe we can make it be able to use standard plugins. How difficult would it be? Thanks so much for your work.
Execute python3 -m fastchat.model.apply_delta --base-model-path E:\\llama-13b-hf --target-model-path D:\\vicuna-13b-delta-v0 --delta-path D:\\13b --low-cpu-mem, then show: Traceback (most recent call last): File "D:\FastChat-main\FastChat-main\fastchat\model\apply_delta.py", line 161, in apply_delta_low_cpu_mem( File "D:\FastChat-main\FastChat-main\fastchat\model\apply_delta.py", line 71, in apply_delta_low_cpu_mem...
