INDEX

Explanations

names of female characters or individuals

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

TV show and network names

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

Personendaten

-0.65

ConstraintMaker

-0.60

.*")]

-0.57

WebElementEntity

-0.54

AutoField

-0.53

numerusform

-0.53

 meriva

-0.52

Autoritní

-0.51

]")]

-0.50

 rospy

-0.48

POSITIVE LOGITS

Demografie

0.35

oredCriteria

0.32

 Wallflower

0.31

تفصیلات

0.31

 CreateTagHelper

0.30

Grease

0.29

Naphthalene

0.29

getOutputStream

0.28

windigkeit

0.28

ucción

0.28

Activations Density 0.027%

names of female characters or individuals

TV show and network names

No Comments

No Known Activations