INDEX

Explanations

references to the concept of "knock" or related terms and names

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 35-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_35/width_16k/average_l0_120

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.35.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

adpleegd

-0.49

SharedDtor

-0.45

 DEFENSE

-0.45

SourceChecksum

-0.44

 faſt

-0.43

stalt

-0.43

 Majefty

-0.42

ſelf

-0.42

éal

-0.42

 aDecoder

-0.41

POSITIVE LOGITS

Kn

0.87

Kno

0.78

KN

0.77

Kn

0.75

kn

0.73

 knock

0.69

Knock

0.67

Kno

0.66

KN

0.66

 Knock

0.66

Activations Density 0.218%

references to the concept of "knock" or related terms and names

No Comments

No Known Activations