INDEX

Explanations

numerical values and mathematical expressions in structured data

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 unequ

-0.98

 buggy

-0.98

minster

-0.95

 wallets

-0.92

quickShipAvailable

-0.92

 Mous

-0.90

 "$:/

-0.90

ollah

-0.88

gebra

-0.88

 Bunny

-0.88

POSITIVE LOGITS

1.32

define

1.30

1.23

ffff

1.22

1.20

dden

1.19

1.18

Activations Density 0.738%

numerical values and mathematical expressions in structured data

No Comments

No Known Activations