INDEX

Explanations

indications of successful function execution or return statements in code

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

return 0; or return nil

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

majánló

-0.92

enablog

-0.91

ロウィン

-0.89

ſſung

-0.87

transQ

-0.87

niſſe

-0.85

ſchaft

-0.85

ésultats

-0.85

 Menſchen

-0.84

 témoig

-0.84

POSITIVE LOGITS

permitAll

0.32

 лёг

0.30

 tranquilo

0.30

NO

0.29

easy

0.28

 easy

0.28

 perfecto

0.28

IN

0.27

NC

0.27

Easy

0.27

Activations Density 0.006%

indications of successful function execution or return statements in code

return 0; or return nil

No Comments

No Known Activations