INDEX

Explanations

specific technical or scientific terms and measurements

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

mathematical and technical 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

BeginContext

-0.75

rungsseite

-0.71

EndContext

-0.70

Personensuche

-0.65

ChildScrollView

-0.63

imanapun

-0.60

OGND

-0.59

expandindo

-0.59

 '\\;'

-0.57

angliski

-0.57

POSITIVE LOGITS

stre

0.39

 bers

0.35

We

0.35

South

0.35

 linear

0.35

バリー

0.35

 trembled

0.35

 Blue

0.34

Sharon

0.34

Blue

0.34

Activations Density 0.132%

specific technical or scientific terms and measurements

mathematical and technical terms

No Comments

No Known Activations