INDEX

Explanations

words related to repairs and maintenance, especially in the context of equipment and vehicles

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Configuration

neuronpedia/gpt2-small__res_scl-ajt/6-res_scl-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

usha

-0.71

 Flavoring

-0.67

ãħĭ

-0.67

arez

-0.65

unin

-0.65

corn

-0.64

ramid

-0.64

soType

-0.64

rencies

-0.62

otine

-0.62

POSITIVE LOGITS

 repair

0.95

 repairs

0.85

 repairing

0.84

repair

0.83

 Kits

0.71

able

0.69

 ashore

0.69

 repaired

0.69

ments

0.67

ability

0.67

Activations Density 5.443%

words related to repairs and maintenance, especially in the context of equipment and vehicles

No Comments

No Known Activations