Pinned Repositories
RWKV-LM
RWKV (pronounced RwaKuv) is an RNN with great LLM performance, which can also be directly trained like a GPT transformer (parallelizable). We are at RWKV-7 "Goose". So it's combining the best of RNN and transformer - great performance, linear time, constant space (no kv-cache), fast training, infinite ctx_len, and free sentence embedding.
inseq
Interpretability for sequence generation models 🐛 🔍
awesome-long-context
DebatingArena
eidf-docs
EIDF Services Documentation
GlyphBERT
HFUTXC-CGS
Classroom Guide System
pretraining-data-packing
SemDeDup
Code for "SemDeDup", a simple method for identifying and removing semantic duplicates from a dataset (data pairs which are semantically similar, but not exactly identical).
StockMarketSummary
MarshtompCS's Repositories
MarshtompCS/DebatingArena
MarshtompCS/GlyphBERT
MarshtompCS/HFUTXC-CGS
Classroom Guide System
MarshtompCS/awesome-long-context
MarshtompCS/eidf-docs
EIDF Services Documentation
MarshtompCS/pretraining-data-packing
MarshtompCS/SemDeDup
Code for "SemDeDup", a simple method for identifying and removing semantic duplicates from a dataset (data pairs which are semantically similar, but not exactly identical).
MarshtompCS/StockMarketSummary