INDEX

Explanations

structured identifiers and categories in a document

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

ValueStyle

-0.54

ніципалі

-0.45

Pautan

-0.42

 carre

-0.42

 IconData

-0.40

yarnpkg

-0.39

getreten

-0.39

createTextNode

-0.38

+:+

-0.38

Nuorodos

-0.38

POSITIVE LOGITS

uxxxx

0.65

出版年

0.47

WithIOException

0.47

opheles

0.45

IsContent

0.45

 mijne

0.43

:]:

0.42

oneofs

0.42

ardonnay

0.41

 kasarigan

0.39

Activations Density 0.507%

structured identifiers and categories in a document

No Comments

No Known Activations