INDEX

Explanations

references to specific roles and actions in societal or political contexts

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

specific technical or foreign words

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

彼

-0.33

shadowRadius

-0.32

av

-0.30

gr

-0.29

sweise

-0.28

Phaser

-0.27

振

-0.27

Based

-0.27

 Zeitpunkt

-0.27

bės

-0.26

POSITIVE LOGITS

Autoritní

0.82

PreferredItem

0.82

tvguidetime

0.75

RenderAtEndOf

0.68

 שוליים

0.66

IUrlHelper

0.66

 propOrder

0.66

 ujednoznacz

0.65

XmlAccessorType

0.64

FieldBuilder

0.64

Activations Density 0.053%

references to specific roles and actions in societal or political contexts

specific technical or foreign words

No Comments

No Known Activations