INDEX

Explanations

references to purpose or intent

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

Negative Logits

 distance

-1.54

 forthcoming

-1.48

rapeut

-1.43

 discrim

-1.41

 times

-1.39

lectric

-1.39

 distant

-1.37

harm

-1.36

 suppressing

-1.34

neut

-1.32

POSITIVE LOGITS

erals

1.85

isters

1.81

ified

1.78

ners

1.77

leys

1.75

neys

1.74

ifiers

1.73

ifications

1.66

ifies

1.63

ves

1.59

Activations Density 0.006%

references to purpose or intent

No Known Activations