INDEX

Explanations

references to specific names or keywords like "Key," "Carpenter," "Zealand," and "Keystone."

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

uates

-0.68

encer

-0.68

 tremend

-0.66

ORK

-0.66

ATIONAL

-0.66

¹

-0.65

 Constantin

-0.64

¶æ

-0.62

BILITY

-0.62

ORED

-0.62

POSITIVE LOGITS

words

0.97

word

0.92

boards

0.92

hart

0.87

ansas

0.85

caps

0.81

 Lime

0.79

hole

0.79

WOR

0.78

ham

0.77

Activations Density 8.126%

references to specific names or keywords like "Key," "Carpenter," "Zealand," and "Keystone."

No Comments

No Known Activations