INDEX

Explanations

specific names and titles related to popular culture, particularly movies and shows

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

RenderAtEndOf

-1.13

rungsseite

-0.88

存于互联网档案馆

-0.80

 فريبيس

-0.75

Personendaten

-0.74

]")]

-0.73

 '\\;'

-0.72

 الرياضيه

-0.71

хьтан

-0.66

 мәкал

-0.65

POSITIVE LOGITS

PhysRevLett

0.50

ifoli

0.43

🅱

0.43

 hate

0.42

elapsed

0.42

yea

0.41

 correction

0.40

◯

0.39

***

0.39

oooooooooooooooo

0.39

Activations Density 0.350%

specific names and titles related to popular culture, particularly movies and shows

No Comments

No Known Activations