INDEX

Explanations

terms related to archival 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

Negative Logits

LookAnd

-0.98

 مرئيه

-0.83

 /\.(

-0.82

principalColumn

-0.80

,"%

-0.79

 فريبيس

-0.77

queryInterface

-0.75

 Cowell

-0.75

 ladders

-0.73

rboles

-0.72

POSITIVE LOGITS

 Archive

1.31

 archive

1.27

Archive

1.23

 archives

1.20

archive

1.19

ARCHIVE

1.06

 Archives

1.03

archives

1.03

Archives

0.90

 archival

0.81

Activations Density 0.005%

terms related to archival content

No Known Activations