INDEX

Explanations

terms related to geographical locations and political events

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

plet

-0.92

rentice

-0.88

ertation

-0.86

psey

-0.83

otom

-0.82

erva

-0.81

olicy

-0.81

lot

-0.81

odder

-0.80

otion

-0.78

POSITIVE LOGITS

ledged

0.90

Ð»

0.85

ties

0.81

cut

0.79

 abbrevi

0.78

itarian

0.77

 Known

0.75

abouts

0.72

cuts

0.71

iary

0.71

Activations Density 9.497%

terms related to geographical locations and political events

No Comments

No Known Activations