INDEX

Explanations

phrases indicating importance or emphasis

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

IntoConstraints

-0.52

хьтан

-0.47

StructEnd

-0.46

 Wikimedijinoj

-0.43

 BoxFit

-0.42

орк

-0.41

phie

-0.41

IUrlHelper

-0.40

ViewFeatures

-0.40

óns

-0.40

POSITIVE LOGITS

AndEndTag

0.54

 חיצוניים

0.50

 незавершена

0.50

 autorytatywna

0.47

 îna

0.45

 dedans

0.42

uğ

0.42

ftagPool

0.41

----</

0.39

OGND

0.39

Activations Density 0.004%

phrases indicating importance or emphasis

No Comments

No Known Activations