INDEX

Explanations

references to moderation in various contexts

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

FPar

-1.49

woke

-1.39

zym

-1.35

cca

-1.33

kte

-1.32

ze

-1.30

ows

-1.29

au

-1.28

ClCompile

-1.28

LETE

-1.27

POSITIVE LOGITS

Īĺ

2.17

 minded

1.66

°

1.59

tees

1.59

minded

1.57

fix

1.56

oon

1.56

ships

1.54

 amounts

1.51

 purposes

1.46

Activations Density 0.015%

references to moderation in various contexts

No Comments

No Known Activations