INDEX

Explanations

any mentions of the term "ObamaCare"

oai_token-act-pair · gpt-3.5-turbo

references to "ObamaCare" and related health terms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

iers

-0.88

ier

-0.88

Lago

-0.83

riv

-0.77

eros

-0.75

ho

-0.75

mere

-0.75

ern

-0.74

ault

-0.73

 Padres

-0.73

POSITIVE LOGITS

ufact

0.98

hovah

0.78

achine

0.78

mble

0.76

 guts

0.66

ments

0.66

ä¸ī

0.65

nesia

0.64

eals

0.62

 heart

0.62

Activations Density 0.038%

any mentions of the term "ObamaCare"

references to "ObamaCare" and related health terms

No Comments

No Known Activations