INDEX

Explanations

specific references to scientific terms or objects, particularly in a structured format

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

mathematical and code abbreviations

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

']").

-0.54

쿵

-0.53

刊

-0.51

burton

-0.50

 basket

-0.50

 whiskers

-0.49

basket

-0.49

helf

-0.49

 handicap

-0.48

boBox

-0.48

POSITIVE LOGITS

出版年

0.61

Enllaces

0.53

 pinulongan

0.45

RTHOOK

0.43

DockStyle

0.42

OGND

0.40

 ligiloj

0.39

setVerticalGroup

0.38

 चीज़ों

0.38

KURZBESCHREIBUNG

0.38

Activations Density 0.224%

specific references to scientific terms or objects, particularly in a structured format

mathematical and code abbreviations

No Comments

No Known Activations