INDEX

Explanations

references to academic publishers and locations

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

references to academic publications and publishers

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

cities and countries

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

 Morav

-0.74

 ModelExpression

-0.73

ритори

-0.69

PreferredItem

-0.64

RenderAtEndOf

-0.64

 فريبيس

-0.62

 fischer

-0.59

ⓧ

-0.59

djangoproject

-0.58

astéro

-0.57

POSITIVE LOGITS

出版年

0.55

New

0.53

USA

0.50

UK

0.50

ambo

0.49

 Royaume

0.49

 Canada

0.48

 London

0.47

NY

0.47

COUVER

0.46

Activations Density 0.051%

references to academic publishers and locations

references to academic publications and publishers

cities and countries

No Comments

No Known Activations