INDEX

Explanations

references to specific tattoo designs, particularly crosses

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

Tattoos, especially crosses

np_acts-logits-general · gemini-2.0-flash

tattoo crosses

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 24-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_24/width_16k/average_l0_73

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.24.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

XDECREF

-0.73

 SwiftUI

-0.68

 ब्रेकडाउन

-0.63

ModelAdmin

-0.62

+#+

-0.57

 binoculars

-0.56

Sortie

-0.56

 noqa

-0.56

 webinars

-0.54

füh

-0.53

POSITIVE LOGITS

 tattoo

1.40

 tattoos

1.27

 Tattoo

1.17

 tattooed

1.10

 inked

1.10

 Tattoos

1.08

ink

1.06

Tattoo

1.05

tattoo

1.04

 tats

1.04

Activations Density 0.083%

references to specific tattoo designs, particularly crosses

Tattoos, especially crosses

tattoo crosses

No Known Activations