INDEX

Explanations

numerical identifiers or cataloging entries

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

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-res/layer_20/width_131k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

IndentedString

-0.65

 EconPapers

-0.63

 ویکی‌پدی

-0.60

 متعلقه

-0.60

UnknownFieldSet

-0.54

 wireType

-0.50

PyExc

-0.50

$_['

-0.49

 ModelExpression

-0.49

MENAFN

-0.49

POSITIVE LOGITS

born

0.45

 born

0.38

ij

0.37

known

0.36

made

0.35

 surla

0.35

 printed

0.35

print

0.35

née

0.34

 perished

0.34

Activations Density 0.022%

numerical identifiers or cataloging entries

No Comments

No Known Activations