INDEX

Explanations

phrases indicating substantial weight or effort

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 12-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_12/width_16k/average_l0_130

Prompts (Dashboard)

24,576 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

writeField

-0.56

AutoField

-0.54

readystatechange

-0.53

 StatelessWidget

-0.53

GPT

-0.52

URLException

-0.51

exitRule

-0.50

 WebDriverWait

-0.50

tlement

-0.50

baarheid

-0.49

POSITIVE LOGITS

 heavy

1.45

heavy

1.42

Heavy

1.41

 Heavy

1.41

 HEAVY

1.37

 heavier

1.09

 heaviest

1.06

 heavily

1.02

 pesados

1.00

 pesada

0.94

Activations Density 0.156%

phrases indicating substantial weight or effort

No Comments

No Known Activations