INDEX

Explanations

phrases expressing legal statements or disclaimers

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Configuration

neuronpedia/gpt2-small__res_scefr-ajt/6-res_scefr-ajt

Prompts (Dashboard)

12,288 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

46,080

Data Type

torch.float32

Hook Point

blocks.6.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

apollo-research/Skylion007-openwebtext-tokenizer-gpt2

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

rea

-0.85

arium

-0.84

seless

-0.83

rex

-0.80

gypt

-0.80

plex

-0.79

gif

-0.74

nas

-0.74

 Orth

-0.72

ln

-0.72

POSITIVE LOGITS

THING

1.50

WHERE

1.14

 particular

1.11

 other

1.07

ONE

1.06

 conceivable

1.05

 significant

1.05

 meaningful

1.01

ones

0.99

 further

0.99

Activations Density 1.233%

phrases expressing legal statements or disclaimers

No Comments

No Known Activations