🔄 In a Training Loop

Giles Thomas

gpjt

4

·

https://www.gilesthomas.com/

AI & ML interests

Doing my best to speedrun 20 years of AI research. YMMV

Recent Activity

updated a model 17 days ago

gpjt/8xa100m40-lore-11-output-only-corrected-smart-init

updated a model 17 days ago

gpjt/8xa100m40-lore-10-both-corrected-smart-init

updated a model 17 days ago

gpjt/8xa100m40-lore-9-new-seed-output-only

View all activity

Organizations

None yet

New activity in AndrewThompson1233/maba-v1-architecture about 1 month ago

Factorized embeddings intervention for GPT-2

#1 opened about 1 month ago by

New activity in gpjt/jax-with-mha-bias-fw-fwedu-5050-DEPRECATED about 1 month ago

Factorized embeddings and non-embedding parameter allocation for GPT-2

#1 opened about 2 months ago by

AndrewThompson1233

New activity in gpjt/openassistant-guanaco-llama2-format about 2 years ago

[bot] Conversion to Parquet

#1 opened about 2 years ago by

parquet-converter

Librarian Bot: Add language metadata for dataset

#2 opened about 2 years ago by