INDEX

Explanations

references to the substring "kn."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 37-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_37/width_16k/average_l0_124

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.37.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

 faſt

-0.46

SharedDtor

-0.46

########.

-0.45

adpleegd

-0.44

oredCriteria

-0.42

تقاوى

-0.42

 aDecoder

-0.41

stalt

-0.41

ſelf

-0.41

şiv

-0.41

POSITIVE LOGITS

Kn

0.88

Kno

0.83

KN

0.83

 knock

0.82

Kn

0.81

Knock

0.80

knock

0.78

kn

0.77

 Knock

0.77

 knocking

0.77

Activations Density 0.344%

references to the substring "kn."

No Comments

No Known Activations