INDEX

Explanations

references to communication channels

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 12-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_12/width_16k/average_l0_82

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.12.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

 sourire

-0.72

ydi

-0.72

 BeautifulSoup

-0.71

 moſt

-0.71

 StatelessWidget

-0.69

 Diſ

-0.67

McK

-0.67

 Reſ

-0.66

 poème

-0.65

 marito

-0.65

POSITIVE LOGITS

 channels

1.77

Channels

1.63

channels

1.59

 channel

1.58

 Channels

1.56

 Channel

1.52

 CHANNEL

1.45

CHANNEL

1.39

Channel

1.37

channel

1.34

Activations Density 0.071%

references to communication channels

No Comments

No Known Activations