INDEX

Explanations

phrases indicating action or prompts to continue engaging with content

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

+#+#

-1.06

WriteTagHelper

-0.79

awtextra

-0.75

Datuak

-0.74

 ***!

-0.74

 defaultstate

-0.74

Referanser

-0.73

 تكبرها

-0.72

onAttach

-0.71

GEBURTS

-0.71

POSITIVE LOGITS

█

0.47

ಗ

0.46

 here

0.46

ước

0.45

 AQUI

0.43

XtraReports

0.43

页面存档备份

0.41

lim

0.41

simo

0.41

সং

0.40

Activations Density 0.003%

phrases indicating action or prompts to continue engaging with content

No Comments

No Known Activations