INDEX

Explanations

references or indicators of statistical data and analyses within the text

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

="@+

-0.48

 tartalomajánló

-0.48

SequentialGroup

-0.45

LookAnd

-0.44

 surla

-0.44

RegressionTest

-0.44

IsMutable

-0.43

parsedMessage

-0.41

 Signalez

-0.40

astéro

-0.39

POSITIVE LOGITS

idać

0.48

hilangan

0.47

colgante

0.46

reszcie

0.44

 dokun

0.43

 móvel

0.41

ypeł

0.41

hidupan

0.40

caneca

0.39

 adecuadamente

0.39

Activations Density 0.234%

references or indicators of statistical data and analyses within the text

No Comments

No Known Activations