INDEX

Explanations

specific legal and medical terminology

oai_token-act-pair · gpt-4o-mini Triggered by @bot

definitions of terms

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 мәкал

-0.53

 ModelExpression

-0.50



-0.50

 insuffisamment

-0.46

 ***!

-0.46

 arşivlendi

-0.45

dflare

-0.44

Tikang

-0.43

 вікісторінку

-0.42

ientôt

-0.42

POSITIVE LOGITS

 OnInit

0.56

 terminology

0.45

 AssemblyProduct

0.44

Etimología

0.44

 Terms

0.44

 Term

0.43

 Glossary

0.43

 ivelany

0.42

 glossary

0.42

 שוליים

0.41

Activations Density 0.082%

specific legal and medical terminology

definitions of terms

No Comments

No Known Activations