FAR.AI
Frontier alignment research to ensure the safe development and deployment of advanced AI systems.
Popular repositories Loading
-
tuned-lens
tuned-lens PublicTools for understanding how transformer predictions are built layer-by-layer
-
-
learned-planner
learned-planner PublicInterpretability tools for recurrent convolutional networks (DRC) that play Sokoban
-
Repositories
Showing 10 of 42 repositories
- s3bucket_exporter Public Forked from molu8bits/s3bucket_exporter
S3 bucket exporter exposing metrics size and number of objects
AlignmentResearch/s3bucket_exporter’s past year of commit activity - learned-planners-stable-baselines3 Public Forked from AlignmentResearch/stable-baselines3
PyTorch version of Stable Baselines, reliable implementations of reinforcement learning algorithms.
AlignmentResearch/learned-planners-stable-baselines3’s past year of commit activity - DeepGEMM Public Forked from deepseek-ai/DeepGEMM
DeepGEMM: clean and efficient FP8 GEMM kernels with fine-grained scaling
AlignmentResearch/DeepGEMM’s past year of commit activity - refusal_direction Public Forked from andyrdt/refusal_direction
Code and results accompanying the paper "Refusal in Language Models Is Mediated by a Single Direction".
AlignmentResearch/refusal_direction’s past year of commit activity