INDEX

Explanations

expressions related to belief and skepticism

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

after consequence or realization

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

KURZBESCHREIBUNG

-0.57

 démocr

-0.54

 colgante

-0.52

 tenue

-0.50

 olympique

-0.49

 característica

-0.48

 informée

-0.48

makeConstraints

-0.48

 clínicos

-0.48

 casera

-0.47

POSITIVE LOGITS

BeginContext

0.40

 uLocal

0.37

 plea

0.35

########.

0.35

fjspx

0.35

Читати

0.33

Dis

0.33

EqualsAnd

0.33

VYMaps

0.33

FS

0.33

Activations Density 0.043%

expressions related to belief and skepticism

after consequence or realization

No Comments

No Known Activations