INDEX

Explanations

phrases indicating superiority or importance of certain items or concepts

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

emerged as, showed as, useful tool

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

 tapasztal

-0.44

 Houſe

-0.44

 feen

-0.42

Datuak

-0.40

ſelf

-0.39

styleUrls

-0.39

 Majefty

-0.38

SharedDtor

-0.38

 dialami

-0.38

ToLeft

-0.37

POSITIVE LOGITS

httphttps

0.48

IsContent

0.44

 ModelExpression

0.40

 uitz

0.39

RetentionPolicy

0.39

Rhestr

0.37

BuildContext

0.36

estimm

0.35

 Neub

0.35

Xna

0.35

Activations Density 0.045%

phrases indicating superiority or importance of certain items or concepts

emerged as, showed as, useful tool

No Comments

No Known Activations