INDEX

Explanations

numerical or coded data representations

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 мәкал

-0.80

országban

-0.70

BeginContext

-0.68

 estekak

-0.65

Personensuche

-0.64

 resourceCulture

-0.64

IndentedString

-0.63

 وتسجيلات

-0.60

 ويكيپيديا

-0.59

seteq

-0.59

POSITIVE LOGITS

chkin

0.56

Παραπομπές

0.55

letal

0.52

Callbacks

0.51

 eles

0.51

terness

0.51

car

0.50

 branch

0.50

ių

0.50

quante

0.49

Activations Density 0.778%

numerical or coded data representations

No Comments

No Known Activations