INDEX

Explanations

references to official announcements or statements

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

INVISIBLE

-0.36

ishan

-0.33

any

-0.32

 smell

-0.32

原始内容存档

-0.32

뮬

-0.31

 anywhere

-0.31

 άν

-0.31

ook

-0.30

 somewhere

-0.30

POSITIVE LOGITS

contentLoaded

0.74

 للاسماء

0.65

 autorytatywna

0.65

 createSlice

0.61

rungsseite

0.58

parsedMessage

0.58

featureID

0.57

UserScript

0.57

newswire

0.56

Tembelea

0.55

Activations Density 0.007%

references to official announcements or statements

No Comments

No Known Activations