INDEX

Explanations

phrases related to communication and interactions about permissions or requests

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

providing assistance or services

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

SharedCtor

-0.73

SharedDtor

-0.58

 >=",

-0.55

onViewCreated

-0.54

GOTREF

-0.51

complexContent

-0.50

__(/*!

-0.49

ngdoc

-0.47

yntaxException

-0.47

POSITIVE LOGITS

 المعيارى

0.46

 Roskov

0.44

 لام

0.44

 OFFIC

0.43

 الدولى

0.39

 Waray

0.39

 Officials

0.39

ThroughAttribute

0.38

 officials

0.38

HtmlAttribute

0.38

Activations Density 0.151%

phrases related to communication and interactions about permissions or requests

providing assistance or services

No Comments

No Known Activations