INDEX

Explanations

negation phrases indicating absence or lack

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

New Auto-Interp

Configuration

ctigges/pythia-70m-deduped__res-sm_processed/1-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.1.hook_resid_post

Hook Layer

Architecture

standard

Context Size

128

Dataset

EleutherAI/the_pile_deduplicated

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

↵

-1.65

 Medic

-1.40

mable

-1.34

ption

-1.31

 millions

-1.31

plet

-1.30

Som

-1.28

 prominently

-1.28

Not

-1.26

 backs

-1.26

POSITIVE LOGITS

wegian

2.77

adren

2.23

ī

2.12

§

2.11

²

2.07

«

2.07

¥

1.90

¼

1.90

Ī

1.88

ª

1.87

Activations Density 0.127%

negation phrases indicating absence or lack

No Comments

No Known Activations