INDEX

Explanations

references to numerical values or rankings

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_21/width_16k/average_l0_70

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

bbene

-0.68

 tả

-0.59

tvguidetime

-0.58

lihood

-0.57

ionario

-0.56

skosten

-0.56

 LayoutInflater

-0.56

Twas

-0.55

 redor

-0.55

sigmoid

-0.54

POSITIVE LOGITS

 numbers

0.91

 NUMBER

0.86

 Number

0.84

 crun

0.83

 Numbers

0.83

 number

0.82

 getNumber

0.80

numbers

0.79

 NUMBERS

0.75

Numbers

0.73

Activations Density 0.129%

references to numerical values or rankings

No Comments

No Known Activations