INDEX

Explanations

phrases emphasizing the significance of certain concepts

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ĥ½

-1.79

Ĥ

-1.78

athing

-1.64

·

-1.63

Ń

-1.61

ı

-1.61

rency

-1.59

 Cities

-1.59

¢

-1.53

ĥ

-1.49

POSITIVE LOGITS

nel

2.02

aliana

1.88

iop

1.74

nell

1.70

binding

1.56

iom

1.56

meal

1.55

ologic

1.49

avis

1.47

ograph

1.46

Activations Density 0.028%

phrases emphasizing the significance of certain concepts

No Comments

No Known Activations