INDEX

Explanations

references to comparison or similarity

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

that of

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-65k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_65k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

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

-0.54

lottes

-0.50

he

-0.48

and

-0.46

New

-0.42

 Major

-0.41

Mor

-0.40

Tag

-0.39

 personal

-0.39

POSITIVE LOGITS

RenderAtEndOf

1.02

 ModelExpression

0.91

DockStyle

0.87

 autorytatywna

0.86

 hers

0.83

 uſed

0.83

Datuak

0.81

 ours

0.80

 المعيارى

0.80

 milik

0.78

Activations Density 0.070%

references to comparison or similarity

that of

No Comments

No Known Activations