INDEX

Explanations

concepts and discussions related to learning

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

 <=",

-0.66

ftagPool

-0.59

 ویکی‌پدی

-0.59

 ModelExpression

-0.57

الحياه

-0.54

 ProtoMessage

-0.54

 deſſen

-0.53

出版年

-0.53

 mxArray

-0.52

 Waſſer

-0.50

POSITIVE LOGITS

 مؤرشف

0.33

 EconPapers

0.31

 RegExp

0.28

usiai

0.28

GEBURTSDATUM

0.28

mulos

0.27

 schedules

0.27

 الرياضيه

0.27

TabbedPane

0.26

 nevoie

0.26

Activations Density 0.010%

concepts and discussions related to learning

No Comments

No Known Activations