INDEX

Explanations

hot topics or issues within a given context

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

ufact

-1.09

 confir

-1.06

ajor

-1.02

uther

-1.00

INAL

-0.96

xus

-0.95

atively

-0.95

ortium

-0.94

acular

-0.94

vous

-0.94

POSITIVE LOGITS

hens

1.14

headed

1.13

 spots

1.13

ness

1.13

bed

1.12

 stove

1.12

ened

1.10

 Chili

1.09

shots

1.09

dogs

1.07

Activations Density 1.261%

hot topics or issues within a given context

No Comments

No Known Activations