INDEX

Explanations

phrases related to political debates and analysis

oai_token-act-pair · gpt-3.5-turbo

references to government, political processes, and political figures

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.7.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

NAD

-0.78

initely

-0.70

 Processing

-0.69

 Variant

-0.69

 Torrent

-0.66

~/

-0.66

iple

-0.65

 Regist

-0.64

 Tags

-0.63

 Storage

-0.63

POSITIVE LOGITS

 ideologically

1.02

 bipartisan

0.99

 populism

0.98

Democrats

0.97

 Democrats

0.95

 populist

0.95

 centrist

0.94

 voters

0.94

 unpopular

0.92

 Republicans

0.92

Activations Density 1.143%

phrases related to political debates and analysis

references to government, political processes, and political figures

No Comments

No Known Activations