INDEX

Explanations

phrases indicating coercion or compulsion

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

Negative Logits

<?

-0.67

verwijspagina

-0.66

zzleHttp

-0.64

AndEndTag

-0.63

GetWindow

-0.63

 Ajoutez

-0.61

yelidikan

-0.61

katapos

-0.60

 thưởng

-0.59

jsxFileName

-0.59

POSITIVE LOGITS

 forced

0.77

 Forced

0.77

Forced

0.77

forced

0.69

 necessitated

0.66

 forcé

0.64

被迫

0.64

 compelled

0.63

 gezwungen

0.62

 mussten

0.61

Activations Density 0.045%

phrases indicating coercion or compulsion

No Known Activations