INDEX

Explanations

references to legal and procedural elements in a context involving authority, decisions, and evaluation of circumstances

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

New Auto-Interp

Configuration

ctigges/pythia-70m-deduped__res-sm_processed/1-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.1.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

breviations

-1.81

à¯

-1.58

reements

-1.54

 tongues

-1.51

 SUCH

-1.51

MENTS

-1.50

ashes

-1.46

jections

-1.44

))$.

-1.43

 reactions

-1.40

POSITIVE LOGITS

 estate

1.68

 govern

1.62

naire

1.54

consin

1.51

oku

1.51

kowski

1.50

 itself

1.50

andom

1.48

isco

1.47

keit

1.42

Activations Density 2.739%

references to legal and procedural elements in a context involving authority, decisions, and evaluation of circumstances

No Comments

No Known Activations