INDEX

Explanations

phrases centered around self-reflection and individual judgment

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

judge for yourself

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

 CreateTagHelper

-0.46

verifyException

-0.44

 createSprite

-0.43

 виправивши

-0.42

rrggbb

-0.38

Biografia

-0.37

يكب

-0.36

 tolua

-0.35

:+:

-0.35

 Wesleyan

-0.34

POSITIVE LOGITS

تقاوى

0.63

 AssemblyCompany

0.43

ImageContext

0.42

DeleteBehavior

0.41

 Putra

0.40

AutoScale

0.40

Perception

0.40

StoreMessageInfo

0.39

 lies

0.39

awtextra

0.39

Activations Density 0.025%

phrases centered around self-reflection and individual judgment

judge for yourself

No Comments

No Known Activations