INDEX

Explanations

programming terms and code-related information

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

 gratification

-0.83

arial

-0.82

 afore

-0.78

 succeeding

-0.78

uate

-0.77

 headache

-0.75

 enclosed

-0.74

 arthritis

-0.73

 Prelude

-0.73

 unpre

-0.73

POSITIVE LOGITS

ITNESS

1.49

INGS

1.45

ITCH

1.41

OW

1.37

OOD

1.34

edge

1.33

atson

1.31

ipe

1.30

atts

1.29

ILD

1.28

Activations Density 11.034%

programming terms and code-related information

No Comments

No Known Activations