INDEX

Explanations

topics related to political figures and social issues

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

bies

-0.93

ares

-0.83

esm

-0.79

 attaches

-0.76

eers

-0.76

otic

-0.75

ð

-0.75

ists

-0.74

 Compan

-0.74

jam

-0.74

POSITIVE LOGITS

 always

1.33

å¥³

1.21

 plenty

1.20

 definitely

1.18

 ALWAYS

1.08

 defin

1.06

 also

1.05

always

1.05

rael

1.03

 ample

1.03

Activations Density 1.012%

topics related to political figures and social issues

No Comments

No Known Activations