INDEX

Explanations

terms related to prosecutorial misconduct and legal proceedings

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

 eins

-0.63

DQ

-0.62

 Werth

-0.59

guri

-0.58

 ROOM

-0.57

Dz

-0.56

Dz

-0.56

рь

-0.55

fml

-0.55

 rooms

-0.54

POSITIVE LOGITS

 prosecute

1.19

 Prosecutor

1.16

 prosecutor

1.03

 prosecutors

1.01

 prosec

1.00

 prosecuting

0.97

secutions

0.95

 prosecuted

0.94

 ExecuteAsync

0.93

 Prosec

0.90

Activations Density 0.012%

terms related to prosecutorial misconduct and legal proceedings

No Known Activations