INDEX

Explanations

references to roles or positions within organizations, particularly in the context of tours or services

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

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-res/layer_20/width_131k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.hook_resid_post

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

 moulin

-0.34

ņi

-0.31

spesies

-0.30

 Dalai

-0.30

 lipca

-0.29

Howdy

-0.29

 démo

-0.29

 Norvège

-0.29

ņas

-0.28

Persistent

-0.28

POSITIVE LOGITS

 chauffeur

0.71

 limousine

0.66

 chauff

0.65

 chauffe

0.63

 taxi

0.60

chauff

0.56

 vehicle

0.55

 transportation

0.55

ousine

0.55

 transports

0.54

Activations Density 0.001%

references to roles or positions within organizations, particularly in the context of tours or services

No Comments

No Known Activations