INDEX

Explanations

names and terms related to North Korea, specifically Kim Jong-un and related issues like ballistic missiles and Korean relations

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

rawdownloadcloneembedreportprint

-0.91

BACK

-0.77

ICAN

-0.76

 Proced

-0.76

Results

-0.75

ATTLE

-0.75

ION

-0.73

ENC

-0.73

WORK

-0.72

ANT

-0.71

POSITIVE LOGITS

sung

0.96

ongyang

0.95

lasses

0.88

 Jong

0.85

stress

0.83

ishi

0.81

Suk

0.78

unta

0.78

enei

0.77

zeb

0.77

Activations Density 6.558%

names and terms related to North Korea, specifically Kim Jong-un and related issues like ballistic missiles and Korean relations

No Comments

No Known Activations