INDEX

Explanations

temporal phrases and indicators suggesting sequences or occurrences

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

nextElement

-0.63

inci

-0.60

Altri

-0.57

 grossesse

-0.57

 sund

-0.56

ISCHE

-0.54

 pleg

-0.54

фициальный

-0.54

nextSibling

-0.54

 Longfellow

-0.53

POSITIVE LOGITS

 للاسماء

0.80

 Bertha

0.74

rceil

0.74

piram

0.73

ECC

0.73

luv

0.73

OGND

0.71

 AssemblyTitle

0.71

GraphicsUnit

0.71

CQ

0.70

Activations Density 0.025%

temporal phrases and indicators suggesting sequences or occurrences

No Comments

No Known Activations