INDEX

Explanations

references to news sources or publication indicators

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

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 conven

-0.45

mobi

-0.45

tafogo

-0.43

mix

-0.41

netinet

-0.40

aş

-0.39

lą

-0.39

kosh

-0.39

etag

-0.39

spunkt

-0.39

POSITIVE LOGITS

 ExecuteAsync

0.84

PRWEB

0.75

Personensuche

0.69

حياته

0.69

});*/

0.68

 Савезне

0.65

HtmlAttribute

0.64

contentLoaded

0.64

'],

0.63

WebVitals

0.63

Activations Density 0.073%

references to news sources or publication indicators

No Comments

No Known Activations