INDEX

Explanations

details or descriptions related to fictional concepts or narratives

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 mosqu

-0.79

 adolesc

-0.69

 exting

-0.69

 horizont

-0.69

 Trinidad

-0.69

 residence

-0.68

 partnerships

-0.62

ãĥīãĥ©ãĤ´ãĥ³

-0.61

 partnership

-0.61

 conclud

-0.61

POSITIVE LOGITS

needed

1.02

really

0.91

âĹ¼

0.90

pires

0.89

else

0.89

said

0.87

you

0.84

ever

0.82

dust

0.79

ebook

0.79

Activations Density 1.753%

details or descriptions related to fictional concepts or narratives

No Comments

No Known Activations