INDEX

Explanations

proper nouns, specifically names of people mentioned in 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

 cref

-0.49

 examples

-0.46

Datuak

-0.44

yves

-0.43

 more

-0.42

 تضيفلها

-0.42

 autorytatywna

-0.41

 nonUne

-0.40

 locking

-0.39

 EXAMPLES

-0.39

POSITIVE LOGITS

 indígen

0.46

 termica

0.45

ESM

0.45

findpost

0.44

SRP

0.44

IsMutable

0.44

ագրություններ

0.44

KEA

0.44

 Anſ

0.43

verifyException

0.43

Activations Density 0.014%

proper nouns, specifically names of people mentioned in the text

No Comments

No Known Activations