INDEX

Explanations

words or phrases containing "hi" or "ahi."

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

Text near bracketed citations

np_acts-logits-general · gemini-2.0-flash

hi elded

np_acts-logits-general · gemini-2.5-flash-lite

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ectoria

-0.44

lm

-0.43

nyt

-0.43

chen

-0.41

 externi

-0.41

 tả

-0.41

dum

-0.41

dia

-0.41

los

-0.40

POSITIVE LOGITS

 endwhile

0.79

 '\\;'

0.79

 Réponses

0.78

GOTREF

0.67

mdi

0.67

ंदीखरीदारी

0.67

UIControlState

0.66

]!='

0.64

+#+

0.63

 BorderRadius

0.63

Activations Density 0.155%

words or phrases containing "hi" or "ahi."

Text near bracketed citations

hi elded

No Comments

No Known Activations