INDEX

Explanations

numerical values and their formatting in 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 @ 13-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_13/width_65k/average_l0_75

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.13.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

 للاسماء

-0.93

featureID

-0.89

AsUp

-0.70

tanleria

-0.70

setVerticalGroup

-0.69

 Wikimedijinoj

-0.68

OwnProperty

-0.65

 noDo

-0.65

 asiático

-0.64

kuuta

-0.63

POSITIVE LOGITS

ileo

0.57

LikeLike

0.54

GetSize

0.50

<?

0.50

bzero

0.49

]='\

0.49

Véxase

0.49

OUGHT

0.48

±

0.48

 Separation

0.47

Activations Density 0.006%

numerical values and their formatting in data representations

No Comments

No Known Activations