Model merging lessons in The Waifu Research Department | allainews.com

Jan. 29, 2024, 4:33 p.m. | Nathan Lambert

Interconnects www.interconnects.ai

When what seems like pure LLM black magic is actually supported by the literature.

literature llm magic merging research

More from www.interconnects.ai / Interconnects

Artifacts Log 1: Announcement, Llama 3 fine-tunes, SOTA reward model, human prompt datasets... 3 days, 20 hours ago | www.interconnects.ai

announcement datasets human llama +5

OpenAI chases Her 6 days ago | www.interconnects.ai

chatgpt her openai society

OpenAI’s Model (behavior) Spec, RLHF transparency, personalization questions 1 week, 3 days ago | www.interconnects.ai

behavior bugs chatgpt effects +7

ChatBotArena: The peoples’ LLM evaluation, the future of evaluation, the incentives of evaluation, and gpt2chatbot 1 week, 5 days ago | www.interconnects.ai

evaluation future incentives llm +3

How RLHF works, part 2: A thin line between useful and lobotomized 2 weeks, 5 days ago | www.interconnects.ai

beyond chat evaluation fine-tuning +5

Phi 3 and Arctic: Outlier LMs are hints 3 weeks ago | www.interconnects.ai

arctic industry llms lms +3

AGI is what you want it to be 3 weeks, 6 days ago | www.interconnects.ai

agi definitions people

WIP Llama 3: Scaling open LLMs 1 month ago | www.interconnects.ai

article llama llama 3 llms +3

Stop "reinventing" everything to solve alignment 1 month ago | www.interconnects.ai

alignment computing everything feedback +7

Software Engineer for AI Training Data (School Specific)

@ G2i Inc | Remote

View on ai-jobs.net

Software Engineer for AI Training Data (Python)

@ G2i Inc | Remote

View on ai-jobs.net

Software Engineer for AI Training Data (Tier 2)

@ G2i Inc | Remote

View on ai-jobs.net

Data Engineer

@ Lemon.io | Remote: Europe, LATAM, Canada, UK, Asia, Oceania

View on ai-jobs.net

Artificial Intelligence – Bioinformatic Expert

@ University of Texas Medical Branch | Galveston, TX

View on ai-jobs.net

Lead Developer (AI)

@ Cere Network | San Francisco, US

View on ai-jobs.net