Burning ray

adarksky
ยท

AI & ML interests

None yet

Recent Activity

updated a model 10 days ago
adarksky/Qwen2.5-0.5B-sft-lora-rel-therapy
published a model 11 days ago
adarksky/Qwen2.5-0.5B-sft-lora-rel-therapy
liked a model 15 days ago
openai/whisper-tiny
View all activity

Organizations

fast.ai community's profile picture Hugging Face Discord Community's profile picture

adarksky's activity

New activity in hexgrad/Kokoro-82M about 1 month ago

Update kokoro.py

#43 opened about 1 month ago by
adarksky
reacted to merve's post with ๐Ÿ”ฅ 2 months ago
view post
Post
2676
small but mighty ๐Ÿ”ฅ
you can fine-tune SmolVLM on an L4 with batch size of 4 and it will only take 16.4 GB VRAM ๐Ÿซฐ๐Ÿป also with gradient accumulation simulated batch size is 16 โœจ
I made a notebook that includes all the goodies: QLoRA, gradient accumulation, gradient checkpointing with explanations on how they work ๐Ÿ’ https://github.com/huggingface/smollm/blob/main/finetuning/Smol_VLM_FT.ipynb