INDEX

Explanations

expressions of physical affection, particularly kissing

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

kissing, kissed, and kisses

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

AllowUser

-0.58

 linho

-0.46

 naquele

-0.42

DataProvider

-0.42

 quadros

-0.41

 fundos

-0.41

 Program

-0.41

 Thebes

-0.40

rativo

-0.39

useState

-0.39

POSITIVE LOGITS

 kiss

1.20

Kiss

1.16

 Kiss

1.14

 kisses

1.11

kiss

1.10

 kissing

1.03

 kissed

1.00

KISS

0.97

 Kisses

0.95

 KISS

0.91

Activations Density 0.005%

expressions of physical affection, particularly kissing

kissing, kissed, and kisses

No Comments

No Known Activations