INDEX

Explanations

phrases that include references to characters or elements from popular culture, particularly in narratives or discussions

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

character descriptions

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

AndEndTag

-0.69

 kasarigan

-0.64

UserScript

-0.59

ConstraintMaker

-0.56

(!__

-0.55

Diwedd

-0.54

onsored

-0.54

ImageContext

-0.53

<unused17>

-0.52

<unused74>

-0.52

POSITIVE LOGITS

 characters

0.48

 character

0.46

 personnage

0.45

Female

0.42

 soldier

0.41

 Characters

0.40

 female

0.40

pic

0.40

female

0.40

 chibi

0.40

Activations Density 0.033%

phrases that include references to characters or elements from popular culture, particularly in narratives or discussions

character descriptions

No Comments

No Known Activations