INDEX

Explanations

programming-related terms and functions for data manipulation

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

PMailer

-0.58

MENAFN

-0.53

 StatelessWidget

-0.48

usda

-0.46

 BorderRadius

-0.45

Boas

-0.44

żesz

-0.44

 informée

-0.44

BeginInit

-0.42

TintMode

-0.42

POSITIVE LOGITS

 maternidade

0.44

المناصب

0.43

 idéia

0.43

BUSY

0.42

asilkan

0.41

 Verhält

0.40

 Geistes

0.39

 irmã

0.38

 sekali

0.37

 paisagem

0.36

Activations Density 0.063%

programming-related terms and functions for data manipulation

No Comments

No Known Activations