INDEX

Explanations

references to extreme importance or significance in a context

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

utmost importance or attention

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.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

SequentialGroup

-0.77

 bagels

-0.59

 panty

-0.55

barrel

-0.54

 graffiti

-0.54

 emancip

-0.52

 Descrip

-0.52

 grenade

-0.52

ectady

-0.52

ffiti

-0.51

POSITIVE LOGITS

 utmost

1.70

 maximum

0.77

 highest

0.75

 extreme

0.68

 Maximum

0.68

 máximo

0.68

 máxima

0.68

 massima

0.66

 massimo

0.66

Maximum

0.65

Activations Density 0.003%

references to extreme importance or significance in a context

utmost importance or attention

No Comments

No Known Activations