GR's picture

5

GR

gr0010

·

AI & ML interests

None yet

Recent Activity

new activity 3 months ago

mattshumer/Reflection-Llama-3.1-70B:I created the Llama-3.1-8B Version

new activity 4 months ago

mattshumer/Reflection-Llama-3.1-70B:The 8B Version Works Better

replied to m-ric's post 4 months ago

🤯 𝗔 𝗻𝗲𝘄 𝟳𝟬𝗕 𝗼𝗽𝗲𝗻-𝘄𝗲𝗶𝗴𝗵𝘁𝘀 𝗟𝗟𝗠 𝗯𝗲𝗮𝘁𝘀 𝗖𝗹𝗮𝘂𝗱𝗲-𝟯.𝟱-𝗦𝗼𝗻𝗻𝗲𝘁 𝗮𝗻𝗱 𝗚𝗣𝗧-𝟰𝗼! @mattshumer, CEO from Hyperwrite AI, had an idea he wanted to try out: why not fine-tune LLMs to always output their thoughts in specific parts, delineated by <thinking> tags? Even better: inside of that, you could nest other sections, to reflect critically on previous output. Let’s name this part <reflection>. Planning is also put in a separate step. He named the method “Reflection tuning” and set out to fine-tune a Llama-3.1-70B with it. Well it turns out, it works mind-boggingly well! 🤯 Reflection-70B beats GPT-4o, Sonnet-3.5, and even the much bigger Llama-3.1-405B! 𝗧𝗟;𝗗𝗥 🥊 This new 70B open-weights model beats GPT-4o, Claude Sonnet, et al. ⏰ 405B in training, coming soon 📚 Report coming next week ⚙️ Uses GlaiveAI synthetic data 🤗 Available on HF! I’m starting an Inference Endpoint right now for this model to give it a spin! Check it out 👉 https://huggingface.co/mattshumer/Reflection-Llama-3.1-70B

View all activity

Organizations

gr0010's activity

New activity in mattshumer/Reflection-Llama-3.1-70B 3 months ago

I created the Llama-3.1-8B Version

#38 opened 4 months ago by

New activity in mattshumer/Reflection-Llama-3.1-70B 4 months ago

The 8B Version Works Better

#44 opened 4 months ago by

replied to m-ric's post 4 months ago

Hi, I made a similar 8B version:
https://huggingface.co/AGI-0/Artificium-llama3.1-8B-001

New activity in featherless-ai/try-this-model 4 months ago

Can you host AGI-0/Artificium-llama3.1-8B-001 ?

#6 opened 4 months ago by

New activity in mattshumer/Reflection-Llama-3.1-70B 4 months ago

Please, 8B version

#8 opened 4 months ago by