INDEX

Explanations

names of people and places

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

bound

-0.76

Bound

-0.74

breakers

-0.74

MM

-0.73

omer

-0.71

breaker

-0.66

breaking

-0.65

leg

-0.65

oming

-0.64

 unequ

-0.64

POSITIVE LOGITS

å§«

1.14

nir

1.02

oulos

1.01

jen

0.99

abwe

0.95

DragonMagazine

0.92

culosis

0.92

enegger

0.92

ruary

0.89

owsky

0.88

Activations Density 6.142%

names of people and places

No Comments

No Known Activations