INDEX

Explanations

information related to agreements, statements, and programs

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

mare

-1.15

UGE

-1.11

liam

-1.11

uge

-1.02

 ceilings

-1.00

 butterflies

-1.00

 nodd

-0.99

ickey

-0.96

UGH

-0.94

 pige

-0.93

POSITIVE LOGITS

 precaution

1.11

displayText

1.11

SW

0.99

SPONSORED

0.98

 bonus

0.95

 consideration

0.94

 Bonus

0.93

uary

0.91

meal

0.89

furt

0.88

Activations Density 0.720%

information related to agreements, statements, and programs

No Comments

No Known Activations