INDEX

Explanations

mentions of the state Nevada

oai_token-act-pair · gpt-3.5-turbo

references to Nevada

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

notation

-0.83

 bearer

-0.83

DragonMagazine

-0.78

 Worker

-0.76

ezvous

-0.74

print

-0.74

 caution

-0.72

xual

-0.71

geist

-0.70

kick

-0.69

POSITIVE LOGITS

ILLE

0.99

ille

0.91

Nev

0.85

isions

0.80

arez

0.79

idia

0.78

ãĤ£

0.76

ince

0.76

Nev

0.76

 Scotia

0.76

Activations Density 0.011%

mentions of the state Nevada

references to Nevada

No Comments

No Known Activations