INDEX

Explanations

terms related to groups of individuals or communities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 1-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_1/width_16k/average_l0_102

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.1.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

kony

-0.63

żdy

-0.61

zeniami

-0.60

Anyone

-0.59

 Anyone

-0.59

 يتيمه

-0.58

Anybody

-0.56

arası

-0.55

 whoſe

-0.55

fera

-0.55

POSITIVE LOGITS

Obrázky

0.80

ьаж

0.65

 ویکی‌پدیا

0.63

 allAfrica

0.59

Hozzáférés

0.59

Билгалдахарш

0.58

 تضيفلها

0.57

Geplaatst

0.56

 peoples

0.56

MessageTagHelper

0.56

Activations Density 0.329%

terms related to groups of individuals or communities

No Comments

No Known Activations