Alexander Borzunov's picture

Alexander Borzunov

borzunov

·

https://github.com/borzunov

AI & ML interests

Making large models more accessible with Petals: https://petals.ml

Organizations

New activity in garage-bAInd/Platypus2-70B-instruct over 2 years ago

Why do you set `use_cache=False`? Removing it will speed up generation

#8 opened over 2 years ago by

New activity in codellama/CodeLlama-34b-Instruct-hf over 2 years ago

Model pads response with newlines up to max_length

#6 opened over 2 years ago by

New activity in tiiuae/falcon-40b over 2 years ago

Revert in-library PR

#90 opened over 2 years ago by

New activity in timdettmers/guanaco-65b over 2 years ago

Add safetensor version

#3 opened over 2 years ago by

New activity in bigscience/bloomz over 2 years ago

A way to inference and fine-tune BLOOMZ-176B from Google Colab or locally

#28 opened almost 3 years ago by

New activity in bigscience/bloom almost 3 years ago

Fine-tune the model?

#46 opened over 3 years ago by

A way to inference and fine-tune BLOOM-176B from Google Colab or locally

#152 opened almost 3 years ago by

What does it take to self-host Bloom? How much money would that cost?

#161 opened almost 3 years ago by

Eats up all RAM + 163GB Swap

#167 opened almost 3 years ago by

LuvIsBadToTheBone

New activity in bigscience/bloomz almost 3 years ago

No longer available, why?

#21 opened about 3 years ago by

New activity in bigscience/bloom-petals almost 3 years ago

Update README.md

#2 opened almost 3 years ago by

New activity in hivemind/gpt-j-6B-8bit almost 3 years ago

hivemind / gpt-j-6B-8bit , How can i use Multiple GPU , I treid using accelerate and also using torch.nn.dataparallel() nothing works out

#11 opened about 3 years ago by

New activity in CALM/arwiki over 3 years ago

Fix `license` metadata

#1 opened over 3 years ago by