INDEX

Explanations

tokens or phrases that indicate high frequency or notable presence in text, particularly those that involve structured documentation or references

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

asiness

-0.54

et

-0.49

NVIC

-0.48

 onCreateView

-0.47

?>">

-0.47

 ?>">

-0.47

 Lick

-0.46

령

-0.45

 rush

-0.44

 informal

-0.44

POSITIVE LOGITS

 للمعارف

0.90

istoitu

0.83

consulté

0.75

 ་་

0.75

 propOrder

0.75

^(@)

0.74

 Houſe

0.73

 purpoſe

0.69

 للاسماء

0.68

 mergeFrom

0.66

Activations Density 0.000%

tokens or phrases that indicate high frequency or notable presence in text, particularly those that involve structured documentation or references

No Known Activations