INDEX

Explanations

numbers and their associations in a structured context, likely pertaining to data or code

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 adaptación

-0.34

 Nerven

-0.33

 meisten

-0.33

ambung

-0.32

 nervioso

-0.30

 sumpay

-0.30

antwo

-0.30

凝土

-0.30

 ettiği

-0.29

 koska

-0.29

POSITIVE LOGITS

MigrationBuilder

0.73

脚注の使い方

0.68

出版年

0.62

 ब्रेकडाउन

0.60

queryInterface

0.60

featureID

0.59

HtmlAttribute

0.56

 Medium

0.55

HasAnnotation

0.54

 المعيارى

0.54

Activations Density 0.008%

numbers and their associations in a structured context, likely pertaining to data or code

No Comments

No Known Activations