INDEX

Explanations

mentions of political partisanship

oai_token-act-pair · gpt-3.5-turbo

references to partisanship and partisan issues

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

uras

-0.84

ternally

-0.84

uran

-0.82

ofi

-0.81

ivia

-0.78

aryn

-0.76

eon

-0.76

iva

-0.75

ivas

-0.73

ea

-0.72

POSITIVE LOGITS

 partisans

1.03

 partisan

1.03

 affiliation

0.84

psy

0.83

 persuasion

0.78

 affili

0.76

 political

0.75

 strugg

0.75

 aggregation

0.74

 propaganda

0.74

Activations Density 0.009%

mentions of political partisanship

references to partisanship and partisan issues

No Comments

No Known Activations