INDEX

Explanations

media references and citations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 13-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_13/width_65k/average_l0_75

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.13.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

انجليز

-0.51

 Picchu

-0.42

 cioè

-0.42

 wajah

-0.42

 δε

-0.40

 henne

-0.40

 terbuka

-0.40

 colis

-0.40

вік

-0.40

 recept

-0.39

POSITIVE LOGITS

expandindo

1.18

 nahilalakip

0.96

 متعلقه

0.94

 дописавши

0.91

NameInMap

0.84

fjspx

0.81

awtextra

0.79

TagMode

0.78

 MainAxisSize

0.78

HtmlAttribute

0.76

Activations Density 0.023%

media references and citations

No Comments

No Known Activations