INDEX

Explanations

phrases related to movement or displacement

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

Predecesor

-0.52

 ویکی‌پدی

-0.50

mær

-0.50

 surla

-0.46

bildēt

-0.45

vateľ

-0.44

 enfans

-0.43

ientras

-0.42

 bertanggung

-0.42

 tartalomajánló

-0.41

POSITIVE LOGITS

躇

0.97

揄

0.53

zzleHttp

0.52

 companyId

0.50

pausal

0.50

parsedMessage

0.48

XtraReports

0.48

checkOut

0.47

الحياه

0.47

Zeneca

0.47

Activations Density 0.080%

phrases related to movement or displacement

No Comments

No Known Activations