INDEX

Explanations

references to famous individuals and historical events

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Pearce

-0.62

 strangers

-0.57

 erect

-0.57

 imaginary

-0.53

 Nile

-0.52

 breeds

-0.52

 IMAGES

-0.52

 Rumble

-0.50

 Mirage

-0.50

 Mongo

-0.49

POSITIVE LOGITS

atural

0.79

otom

0.74

cream

0.72

ules

0.72

kowski

0.71

pool

0.70

ette

0.70

creen

0.70

ellen

0.69

lay

0.69

Activations Density 6.848%

references to famous individuals and historical events

No Comments

No Known Activations