INDEX

Explanations

mentions of prominent political figures

oai_token-act-pair · gpt-3.5-turbo

references to political figures and governmental structures

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

staking

-0.55

DragonMagazine

-0.54

 orally

-0.48

 Kejriwal

-0.47

quickShipAvailable

-0.47

earchers

-0.47

:=

-0.45

 Editor

-0.44

 google

-0.43

 Reporter

-0.43

POSITIVE LOGITS

wealth

0.65

ĸ

0.48

ashington

0.47

°

0.46

 federation

0.46

ļ

0.46

 welfare

0.46

iphate

0.45

%.

0.45

 workforce

0.45

Activations Density 1.372%

mentions of prominent political figures

references to political figures and governmental structures

No Comments

No Known Activations