INDEX

Explanations

references to coefficients and variable storage in mathematical or programming contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Jeografia

-0.63

IndentedString

-0.59

出版年

-0.56

parsedMessage

-0.54

IVEREF

-0.54

 sqlalchemy

-0.52

styleType

-0.51

ConstraintMaker

-0.50

 betweenstory

-0.49

akaian

-0.48

POSITIVE LOGITS

:✨

0.38

INSEE

0.33

 Span

0.32

Autoritní

0.32

 الاطلاع

0.32

:]:

0.31

Superficie

0.30

 Benitez

0.30

 Mayfield

0.30

%%%

0.30

Activations Density 0.022%

references to coefficients and variable storage in mathematical or programming contexts

No Comments

No Known Activations