INDEX

Explanations

tags or markers indicating the beginning of sections or important content

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

 propOrder

-0.36

 pauvreté

-0.33

/*

-0.33

égalité

-0.32

であろう

-0.30

ModelMap

-0.30

classnames

-0.30

coffret

-0.30

jface

-0.29

BrowserModule

-0.29

POSITIVE LOGITS

хьтан

0.64

 kasarigan

0.63

rungsseite

0.61

tagHelperRunner

0.60

تقاوى

0.59

StructEnd

0.57

 censiti

0.57

الحياه

0.57

OGND

0.56

ьажоргаш

0.56

Activations Density 0.000%

tags or markers indicating the beginning of sections or important content

No Comments

No Known Activations