INDEX

Explanations

text related to societal issues, politics, criticism, and observations

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

————————

-0.39

ĵĺ

-0.36

 Belg

-0.35

 supers

-0.34

Tele

-0.34

Construction

-0.33

 additive

-0.33

Effective

-0.33

Mill

-0.33

 Reynolds

-0.33

POSITIVE LOGITS

etooth

0.53

ohm

0.46

miah

0.45

isoft

0.45

umbai

0.44

essor

0.44

peria

0.43

foundland

0.43

ruary

0.43

iflower

0.42

Activations Density 36.040%

text related to societal issues, politics, criticism, and observations

No Comments

No Known Activations