INDEX

Explanations

key names and titles associated with places and people

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 11-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_11/width_16k/average_l0_80

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.11.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

Personensuche

-0.61

org

-0.56

 estekak

-0.55

pat

-0.54

RON

-0.53

lecht

-0.53

Попис

-0.53

 пунктов

-0.52

 TestBed

-0.52

міністра

-0.51

POSITIVE LOGITS

 AssemblyCulture

0.68

ImageContext

0.60

 frattempo

0.60

Efq

0.58

 ſever

0.57

XmlAccessorType

0.56

 auffi

0.55

 iſt

0.54

}\|

0.54

 Majefty

0.53

Activations Density 0.106%

key names and titles associated with places and people

No Comments

No Known Activations