INDEX

Explanations

proper nouns related to individuals and their titles or roles

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

ⓧ

-0.65

majánló

-0.62

 Chwiliwch

-0.62

PostInfinity

-0.61

 BoxFit

-0.60

adaptiveStyles

-0.60

 AssemblyCulture

-0.60

 للمعارف

-0.59

exitRule

-0.59

tvguidetime

-0.59

POSITIVE LOGITS

 director

0.41

Director

0.40

0.39

 Director

0.38

CEO

0.36

director

0.35

 attorney

0.31

Steven

0.31

 DIRECTOR

0.31

CEO

0.30

Activations Density 0.140%

proper nouns related to individuals and their titles or roles

No Comments

No Known Activations