INDEX

Explanations

phrases that indicate the need for communication or inquiries

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

requesting contact

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

slidesToShow

-0.59

 BorderSide

-0.58

 Италијани

-0.58

 rootReducer

-0.58

didSet

-0.55

 vägen

-0.54

 céré

-0.54

 newUser

-0.53

 avoient

-0.53

placer

-0.52

POSITIVE LOGITS

 contact

0.81

contact

0.79

Contact

0.79

 Contact

0.72

 contacting

0.67

 CONTACT

0.66

CONTACT

0.64

 contacted

0.63

 contacts

0.59

contacts

0.59

Activations Density 0.053%

phrases that indicate the need for communication or inquiries

requesting contact

No Comments

No Known Activations