INDEX

Explanations

references to Australia and its regions

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

New Auto-Interp

Configuration

ctigges/pythia-70m-deduped__res-sm_processed/2-res-sm

Prompts (Dashboard)

32,768 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

32,768

Data Type

torch.float32

Hook Name

blocks.2.hook_resid_post

Hook Layer

Architecture

standard

Context Size

128

Dataset

EleutherAI/the_pile_deduplicated

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Ĥ¬

-1.83

Ħ

-1.71

ī

-1.68

 None

-1.63

aggio

-1.58

 something

-1.49

¹

-1.49

§

-1.47

 Ã§a

-1.47

ambda

-1.46

POSITIVE LOGITS

bour

1.81

ctica

1.62

selves

1.61

vist

1.55

 Standards

1.52

ricular

1.48

aurus

1.46

ureus

1.46

ual

1.44

 standards

1.43

Activations Density 0.162%

references to Australia and its regions

No Comments

No Known Activations