Are there are any guides for lm_eval??
#5
by
minjae0047
- opened
I want this model to evaluate in other tasks.
Are there some codes or commands to run in lm_eval??
Thank you
lm_eval should be directly usable with most HF models. You can try something like
accelerate launch -m lm_eval --model hf --model_args pretrained=TRI-ML/mamba-7b-rw --tasks (task-name-here)
sedrickkeh
changed discussion status to
closed