INDEX

Explanations

proper nouns related to political figures, sports teams, and locations

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ACTED

-0.80

Lt

-0.78

Redd

-0.73

Spy

-0.72

UGC

-0.71

CG

-0.70

ulhu

-0.68

slave

-0.67

HI

-0.66

Si

-0.66

POSITIVE LOGITS

 Barron

0.96

otyp

0.75

alon

0.75

asso

0.75

abad

0.75

cloth

0.74

agus

0.71

xual

0.70

mares

0.70

agraph

0.70

Activations Density 0.263%

proper nouns related to political figures, sports teams, and locations

No Comments

No Known Activations