INDEX

Explanations

references to abstract concepts and fundamental ideas 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

Not in Any Lists

No Comments

Negative Logits

 endforeach

-0.65

 tait

-0.58

 fufficient

-0.58

 fhort

-0.58

écnicas

-0.57

 neceff

-0.56

 patchwork

-0.55

RenderAtEndOf

-0.54

 vivre

-0.52

 bustling

-0.52

POSITIVE LOGITS

 abstrak

0.68

yntaxException

0.67

 abstract

0.66

ArrowToggle

0.64

 Abstra

0.62

 betweenstory

0.62

 abstra

0.61

abstract

0.60

 abstrato

0.60

ligg

0.55

Activations Density 0.139%

references to abstract concepts and fundamental ideas in various contexts

No Comments

No Known Activations