INDEX

Explanations

references to narratives or descriptions of experiences

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

-2.28

»¿

-2.21

Ń

-2.17

·

-2.03

®

-1.97

¸

-1.95

¢

-1.95

Ħ

-1.91

³

-1.89

ľĵ

-1.88

POSITIVE LOGITS

 respect

1.56

 DISCLAIM

1.53

*:

1.50

 regard

1.46

1.44

:</

1.43

 willingness

1.43

NON

1.43

 humility

1.42

har

1.42

Activations Density 4.477%

references to narratives or descriptions of experiences

No Comments

No Known Activations