INDEX

Explanations

identifiers or references related to experimental data or results

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

Datuak

-0.82

 RIPRODUZIONE

-0.72

 للمعارف

-0.71

 dévelo

-0.69

⇑

-0.69

لاثة

-0.65

 igång

-0.65

 coussin

-0.63

guantes

-0.60

cámara

-0.60

POSITIVE LOGITS

 Hanno

0.54

enciaga

0.51

 Mila

0.51

 Berto

0.51

cle

0.51

 Yoko

0.51

AFM

0.49

 Beren

0.49

iastes

0.48

 Riverside

0.48

Activations Density 0.002%

identifiers or references related to experimental data or results

No Comments

No Known Activations

identifiers or references related to experimental data or results

No Comments

No Known Activations