INDEX

Explanations

interrogative questions and statements involving intentions or actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

bootstrapcdn

-0.52

 fédé

-0.46

 możli

-0.45

sizeCache

-0.44

 zijne

-0.43

 nødvendig

-0.43

 encantó

-0.42

 terecht

-0.41

pulseira

-0.41

 toată

-0.41

POSITIVE LOGITS

出版年

0.47

windowFixed

0.44

WireFormatLite

0.44

abestanden

0.41

MessageTagHelper

0.40

tou

0.39

ご覧

0.39

adaptiveStyles

0.39

 kane

0.39

aarrggbb

0.38

Activations Density 0.569%

interrogative questions and statements involving intentions or actions

No Comments

No Known Activations