INDEX

Explanations

references to skirts and related clothing items

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

skirt

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

Nox

-0.56

Max

-0.50

 Benef

-0.46

romolecules

-0.46

almaz

-0.45

Jax

-0.45

 Dina

-0.45

 vola

-0.45

 Preven

-0.45

MAX

-0.45

POSITIVE LOGITS

 Skirt

1.16

 skirt

1.15

 skirts

0.98

skirt

0.86

 falda

0.84

 jupe

0.84

skirts

0.84

 faldas

0.74

スカート

0.73

 etek

0.66

Activations Density 0.003%

references to skirts and related clothing items

skirt

No Comments

No Known Activations