INDEX

Explanations

numerical values related to efficiency and performance

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

Personensuche

-0.43

centralwidget

-0.41

cephal

-0.39

pflich

-0.38

]]:

-0.38

صبحت

-0.37

Hyp

-0.37

chter

-0.36

Grants

-0.35

Hyp

-0.35

POSITIVE LOGITS

IsMutable

0.55

webElementXpaths

0.55

ScopeManager

0.53

 ligiloj

0.53

 Italijanski

0.52

 protoimpl

0.49

 myſelf

0.46

parsedMessage

0.45

 ſta

0.45

ſelves

0.44

Activations Density 0.013%

numerical values related to efficiency and performance

No Comments

No Known Activations