INDEX

Explanations

phrases expressing strong opinions or feedback

oai_token-act-pair · gpt-4o-mini

New Auto-Interp

Configuration

chanind/gemma-2-2b-batch-topk-matryoshka-saes-w-32k-l0-40/standard/blocks.12.hook_resid_post

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

32,768

Data Type

float32

Hook Name

blocks.12.hook_resid_post

Architecture

jumprelu

Context Size

1,024

Dataset

chanind/pile-uncopyrighted-gemma-1024-abbrv-1B

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 petals

-0.56

 Wheatley

-0.52

 onOptions

-0.52

 SUDOC

-0.51

前に

-0.50

 seno

-0.50

*}$

-0.50

 верно

-0.50

 iunie

-0.49

uParam

-0.49

POSITIVE LOGITS

parsedMessage

1.19

 تضيفلها

0.91

 nonatomic

0.84

nonatomic

0.84

messageInfo

0.77

LookAnd

0.73

RegistryLite

0.72

脚注の使い方

0.72

Còn

0.72

 Administrativna

0.71

Activations Density 0.002%

phrases expressing strong opinions or feedback

No Comments

No Known Activations