INDEX

Explanations

references to political figures, government policies, and historical 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

 concess

-0.68

 captcha

-0.67

SELECT

-0.62

 Norn

-0.62

 Completed

-0.62

ovember

-0.61

angan

-0.61

 CrossRef

-0.60

amaz

-0.59

 discharge

-0.57

POSITIVE LOGITS

ervatives

1.20

ervative

1.13

 ideology

1.01

 liberals

0.93

 ideologies

0.93

ocracy

0.92

ocrats

0.92

 liberalism

0.91

 hegemony

0.90

 neocons

0.88

Activations Density 19.789%

references to political figures, government policies, and historical events

No Comments

No Known Activations