INDEX

Explanations

references to various objects in a document

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-mlp/layer_6/width_131k/average_l0_72

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.6.hook_mlp_out

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

 objects

-1.62

objects

-1.41

 Objects

-1.41

Objects

-1.16

 objetos

-1.02

 objets

-0.98

 Objekte

-0.91

 oggetti

-0.88

 Objek

-0.82

objetos

-0.81

POSITIVE LOGITS

 comigo

0.44

 AssemblyCompany

0.43

 conmigo

0.43

+#+

0.42

 chắn

0.42

Filmografie

0.41

ंदीखरीदारी

0.40

 disposición

0.39

LookAnd

0.39

 oczy

0.38

Activations Density 0.005%

references to various objects in a document

No Comments

No Known Activations