INDEX

Explanations

references to correlative or related terms in various contexts

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

 Theſe

-1.49

 Anſ

-1.25

 doubtnut

-1.25

 myſelf

-1.25

 Houſe

-1.24

 itſelf

-1.24

Efq

-1.21

 ſhould

-1.20

 whoſe

-1.17

 ſhall

-1.16

POSITIVE LOGITS

Cor

1.61

Cor

1.45

cor

1.39

cor

1.35

COR

1.19

COR

1.07

 Corbett

1.02

 Correa

1.00

com

0.89

 Corvette

0.88

Activations Density 0.019%

references to correlative or related terms in various contexts

No Known Activations