INDEX

Explanations

references to connection or networking terms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 4-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_4/width_16k/average_l0_124

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.4.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

Vader

-0.68

 UIImage

-0.68

aVar

-0.63

ần

-0.63

 Maison

-0.62

krb

-0.62

mäßigen

-0.61

 authService

-0.60

etary

-0.60

RICAL

-0.60

POSITIVE LOGITS

 Connect

1.78

Connect

1.74

connect

1.66

 CONNECT

1.53

CONNECT

1.47

 connect

1.41

 connects

1.15

connects

1.14

Disconnect

1.09

Connecting

1.06

Activations Density 0.182%

references to connection or networking terms

No Comments

No Known Activations