Yu Zhao

yuzhaouoe's activity

• All
• Models
• Datasets
• Spaces
• Papers
• Collections
• Community
• Posts
• Upvotes
• Likes

updated a collection about 2 months ago

SAE-Based Representation Engineering

Collection

Steering Knowledge Selection Behaviours in LLMs via SAE-Based Representation Engineering github.com/yuzhaouoe/SAE-based-representation-engineering • 3 items • Updated Nov 1

commented 2 papers about 2 months ago

Steering Knowledge Selection Behaviours in LLMs via SAE-Based Representation Engineering

Paper • 2410.15999 • Published Oct 21 • 19 •

3

Analysing the Residual Stream of Language Models Under Knowledge Conflicts

Paper • 2410.16090 • Published Oct 21 • 7 •

2

upvoted a paper about 2 months ago

DeCoRe: Decoding by Contrasting Retrieval Heads to Mitigate Hallucinations

Paper • 2410.18860 • Published Oct 24 • 9

liked 2 models about 2 months ago

HIT-TMG/KaLM-embedding-multilingual-mini-v1

Sentence Similarity • Updated Nov 1 • 1.92k • 13

HIT-TMG/KaLM-embedding-multilingual-mini-instruct-v1

Sentence Similarity • Updated Nov 12 • 36.4k • 20

commented a paper about 2 months ago

Steering Knowledge Selection Behaviours in LLMs via SAE-Based Representation Engineering

Paper • 2410.15999 • Published Oct 21 • 19 •

3

upvoted 2 papers about 2 months ago

Analysing the Residual Stream of Language Models Under Knowledge Conflicts

Paper • 2410.16090 • Published Oct 21 • 7

Steering Knowledge Selection Behaviours in LLMs via SAE-Based Representation Engineering

Paper • 2410.15999 • Published Oct 21 • 19

updated a model 2 months ago

yuzhaouoe/Llama2-7b-SAE

Updated Oct 9

upvoted a paper 3 months ago

Gemma Scope: Open Sparse Autoencoders Everywhere All At Once on Gemma 2

Paper • 2408.05147 • Published Aug 9 • 38

updated a collection 4 months ago

Pre-Trianing Data Packing

Collection

[ACL'24] Analysing the Impact of Sequence Composition on Language Model Pre-Training. https://github.com/yuzhaouoe/pretraining-data-packing • 9 items • Updated Sep 6