INDEX

Explanations

sections or components of a detailed report or analysis

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

mixed-language content or symbols

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

ásra

-0.24

 senior

-0.23

 aktivitet

-0.23

expandindo

-0.22

Hi

-0.22

 пользу

-0.21

 expectativas

-0.21

 Senior

-0.21

 zagro

-0.21

agaimana

-0.21

POSITIVE LOGITS

 الحره

0.90

 nakalista

0.87

:✨

0.84

ViewFeatures

0.79

rungsseite

0.77

istoitu

0.77

IsContent

0.72

Hochspringen

0.70

Aktualisiert

0.68

+:+

0.68

Activations Density 0.103%

sections or components of a detailed report or analysis

mixed-language content or symbols

No Comments

No Known Activations