INDEX

Explanations

the word "Will" indicating future actions or inquiries

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

New Auto-Interp

Configuration

ctigges/pythia-70m-deduped__mlp-sm_processed/0-mlp-sm

Prompts (Dashboard)

32,768 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

32,768

Data Type

torch.float32

Hook Name

blocks.0.hook_mlp_out

Hook Layer

Architecture

standard

Context Size

128

Dataset

EleutherAI/the_pile_deduplicated

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ANC

-1.62

ITED

-1.60

orting

-1.52

INCLUDING

-1.51

eah

-1.50

OPE

-1.50

)):

-1.48

upal

-1.42

phrine

-1.41

ATED

-1.38

POSITIVE LOGITS

zier

1.77

avi

1.66

herty

1.63

isters

1.56

iei

1.52

ister

1.50

ough

1.49

iest

1.49

power

1.43

caster

1.42

Activations Density 0.042%

the word "Will" indicating future actions or inquiries

No Comments

No Known Activations