INDEX

Explanations

references to visual content, particularly images or photos

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

images or posts

np_acts-logits-general · gemini-2.0-flash

photos and pictures

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

GraphicsUnit

-0.71

SuppressMessage

-0.61

SharedCtor

-0.57

internalType

-0.56

WebServlet

-0.51

inheritDoc

-0.50

ankind

-0.47

ftagPool

-0.47

fromnode

-0.46

spender

-0.45

POSITIVE LOGITS

 photos

3.05

 pictures

2.96

 photographs

2.77

 photograph

2.62

 photo

2.60

 images

2.58

 pics

2.49

 picture

2.44

 Photos

2.37

 fotos

2.29

Activations Density 0.976%

references to visual content, particularly images or photos

images or posts

photos and pictures

No Comments

No Known Activations