Robert Kirk's picture

23 3 1

Robert Kirk

robkirk

https://robertkirk.github.io/

AI & ML interests

AI Alignment and Safety, Generalisation, RLHF, LLMs

Recent Activity

updated a model 4 days ago

LLM-GAT/llama3_gd_lora-64-16_beta-2p0_bs-32_lr-3e-04_checkpoint-8

New activity 4 days ago

LLM-GAT/llama3_gd_lora-64-16_beta-2p0_bs-32_lr-3e-04_checkpoint-8:Update adapter_config.json

updated a model 4 days ago

LLM-GAT/llama3_gd_lora-64-16_beta-2p0_bs-32_lr-3e-04_checkpoint-7

View all activity

Organizations

robkirk's activity

upvoted 2 papers 11 months ago

Sleeper Agents: Training Deceptive LLMs that Persist Through Safety Training

Paper • 2401.05566 • Published Jan 10 • 26

Mechanistically analyzing the effects of fine-tuning on procedurally defined tasks

Paper • 2311.12786 • Published Nov 21, 2023 • 2

upvoted a paper about 1 year ago

Understanding the Effects of RLHF on LLM Generalisation and Diversity

Paper • 2310.06452 • Published Oct 10, 2023 • 2