INDEX

Explanations

phrases that discuss the organization or arrangement of items or ideas

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

laying out information

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.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

 وتسجيلات

-0.58

écnicas

-0.43

 незавершена

-0.42

 appartiennent

-0.42

esserung

-0.42

 belonged

-0.42

/*

-0.41

 belong

-0.41

isation

-0.41

Boletín

-0.41

POSITIVE LOGITS

 layout

1.02

 Layout

1.01

 layouts

0.90

 LAYOUT

0.89

Layout

0.88

ayout

0.82

layout

0.80

Layouts

0.75

 getLayout

0.74

LAYOUT

0.70

Activations Density 0.005%

phrases that discuss the organization or arrangement of items or ideas

laying out information

No Comments

No Known Activations