INDEX

Explanations

mentions of notable names and locations in articles, possibly related to journalism or literature

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

atio

-0.74

iple

-0.73

ensical

-0.72

ary

-0.72

omen

-0.71

issance

-0.70

onym

-0.68

¬¼

-0.68

 experien

-0.67

aled

-0.66

POSITIVE LOGITS

etc

1.09

etc

0.89

tein

0.82

 Burg

0.70

 Gang

0.70

 including

0.69

Magikarp

0.69

Downloadha

0.65

 preferably

0.65

 which

0.64

Activations Density 2.720%

mentions of notable names and locations in articles, possibly related to journalism or literature

No Comments

No Known Activations