INDEX

Explanations

phrases that indicate recommendations or endorsements

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

 Kün

-0.63

 AssemblyCompany

-0.60

Quoi

-0.60

ولايات

-0.57

ajemy

-0.57

chin

-0.50

 gantung

-0.50

WriteAttribute

-0.49

xture

-0.49

llung

-0.49

POSITIVE LOGITS

 فريبيس

0.64

qrstuvwxyz

0.63

voyez

0.54

Personensuche

0.53

 betweenstory

0.52

 assoluto

0.52

ctory

0.50

skechers

0.50

BURGH

0.50

cosis

0.49

Activations Density 0.009%

phrases that indicate recommendations or endorsements

No Comments

No Known Activations