INDEX

Explanations

references to news and data safety in various contexts

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

UrlResolution

-0.55

 فريبيس

-0.53

 cremation

-0.45

parsedMessage

-0.44

ftagPool

-0.44

 acompañado

-0.44

 sostitu

-0.41

الدراسه

-0.41

 <>",

-0.41

 accesibles

-0.41

POSITIVE LOGITS

 EconPapers

0.44

ViewFeatures

0.42

SharedDtor

0.42

WebElementEntity

0.41

 "$@"

0.41

قایناق‌لار

0.40

expandindo

0.36

PMailer

0.35

 Issus

0.35

 ngoài

0.34

Activations Density 0.206%

references to news and data safety in various contexts

No Comments

No Known Activations