Spaces:

smol-course
/

leaderboard

Running

Update submissions.json

by sanjay-saatyaki - opened Sep 16

←

Sep 16

with T4 Colab GPU. Cant train a model without LORA.
Model trained with LORA is not able to validate with lighteval vllm. So Using accelerate

!lighteval accelerate "model_name=sanjay-saatyaki/smol-train,dtype=float16" "lighteval|gsm8k|0|0" --push-to-hub --results-org sanjay-saatyaki

Sep 16

Try merging lora adapter with the original weights

a smol course org Sep 26

Thanks. This is published!

burtenshaw changed pull request status to closed Sep 26

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment