Shivaen Ramshetty's picture

5 65

Shivaen Ramshetty

shivr

·

sramshetty

AI & ML interests

NLP, CV, Multimodal

Recent Activity

upvoted a paper about 1 month ago

Softpick: No Attention Sink, No Massive Activations with Rectified Softmax

upvoted a paper 12 months ago

Tokenization Falling Short: The Curse of Tokenization

upvoted a paper 12 months ago

TroL: Traversal of Layers for Large Language and Vision Models

View all activity

Organizations

shivr's activity

commented 4 papers about 1 year ago

The Unreasonable Ineffectiveness of the Deeper Layers

Paper • 2403.17887 • Published Mar 26, 2024 • 81 •

The Unreasonable Ineffectiveness of the Deeper Layers

Paper • 2403.17887 • Published Mar 26, 2024 • 81 •

ShortGPT: Layers in Large Language Models are More Redundant Than You Expect

Paper • 2403.03853 • Published Mar 6, 2024 • 65 •

ShortGPT: Layers in Large Language Models are More Redundant Than You Expect

Paper • 2403.03853 • Published Mar 6, 2024 • 65 •

New activity in shivr/gpt2-xl_local-narratives-reduced-overlap_lora over 1 year ago

Librarian Bot: Add base_model information to model

#1 opened over 1 year ago by

New activity in shivr/gpt2-xl_grit_and_local-narratives_lora over 1 year ago

Librarian Bot: Add base_model information to model

#1 opened over 1 year ago by