INDEX

Explanations

phrases related to "roll" or "rolling."

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

 LORD

-1.59

 Parliament

-1.59

 pleasure

-1.55

>()

-1.45

usterity

-1.41

 organised

-1.40

/>

-1.39

rapy

-1.39

 indeed

-1.37

).](

-1.36

POSITIVE LOGITS

ups

1.99

hart

1.73

tees

1.72

ably

1.68

uple

1.67

iom

1.63

icking

1.60

backs

1.59

maker

1.59

able

1.57

Activations Density 0.038%

phrases related to "roll" or "rolling."

No Comments

No Known Activations