INDEX

Explanations

the word "meg" or variations of "meg" as it pertains to measurement or names

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

ãģį

-0.76

 Grounds

-0.73

xual

-0.72

 Paradox

-0.72

eer

-0.66

âķĲâķĲ

-0.66

velt

-0.65

 Sense

-0.65

 condition

-0.64

 humour

-0.64

POSITIVE LOGITS

abytes

1.38

abyte

1.37

awatts

1.18

apixel

1.07

atron

1.00

apolis

0.96

aton

0.93

acity

0.92

atis

0.92

abases

0.90

Activations Density 6.654%

the word "meg" or variations of "meg" as it pertains to measurement or names

No Comments

No Known Activations