INDEX

Explanations

references to images and photographs

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

image, photo, picture

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

setopt

-0.53

stdc

-0.47

HtmlAttribute

-0.42

AddTagHelper

-0.40

AlterField

-0.39

rrggbb

-0.38

abstractmethod

-0.36

DoubleQuotes

-0.36

+:+

-0.35

 będ

-0.33

POSITIVE LOGITS

 pics

0.62

 photographs

0.61

LabelTagHelper

0.61

 fotos

0.60

 pictures

0.60

 photos

0.60

 visuals

0.60

 visuales

0.57

 photography

0.56

Photographs

0.54

Activations Density 0.072%

references to images and photographs

image, photo, picture

No Comments

No Known Activations