INDEX

Explanations

varied instances of quotation marks or special characters in the document

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

onna

-1.61

like

-1.42

 times

-1.40

shot

-1.31

 like

-1.30

 below

-1.30

 normally

-1.28

 donate

-1.27

 shoulders

-1.26

 nightmare

-1.25

POSITIVE LOGITS

idone

1.63

LICENSE

1.60

protobuf

1.56

heets

1.48

uginosa

1.46

EventListener

1.45

udad

1.41

idine

1.39

furt

1.38

hdad

1.38

Activations Density 0.057%

varied instances of quotation marks or special characters in the document

No Known Activations