INDEX

Explanations

titles, names, and specific attributes related to individuals and their roles

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Configuration

ctigges/pythia-70m-deduped__res-sm_processed/2-res-sm

Prompts (Dashboard)

32,768 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

32,768

Data Type

torch.float32

Hook Name

blocks.2.hook_resid_post

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

 analogy

-1.73

itate

-1.64

âĢĲ

-1.62

rization

-1.58

ificial

-1.50

tbl

-1.49

thouse

-1.49

ismo

-1.49

endas

-1.48

idades

-1.45

POSITIVE LOGITS

ĻĤ

3.12

Į

2.75

ĨĴ

2.54

¦

2.51

²

2.50

ľĵ

2.49

Īĺ

2.49

ĺ

2.39

©

2.38

Ĺ

2.36

Activations Density 0.191%

titles, names, and specific attributes related to individuals and their roles

No Comments

No Known Activations