INDEX

Explanations

mentions of specific states in the context of political events or projects

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

eners

-1.10

enegger

-0.93

colo

-0.87

eric

-0.82

ebook

-0.82

etting

-0.79

enance

-0.79

 Gutenberg

-0.77

advertisement

-0.76

hester

-0.73

POSITIVE LOGITS

 Anchorage

1.21

 Airlines

1.15

 Springs

0.93

aska

0.92

 Alaska

0.90

 Native

0.86

ħĭ

0.85

atel

0.83

 Refuge

0.82

 Permanent

0.81

Activations Density 4.154%

mentions of specific states in the context of political events or projects

No Comments

No Known Activations