INDEX

Explanations

mentions of the name "Donald Trump"

oai_token-act-pair · gpt-3.5-turbo

mentions of Donald Trump

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

 peer

-0.67

 otherwise

-0.66

 session

-0.65

 such

-0.65

 gear

-0.64

 squad

-0.63

ams

-0.62

 similarly

-0.62

 labs

-0.61

 even

-0.61

POSITIVE LOGITS

Donald

3.87

donald

2.19

 Donald

2.15

Trump

1.94

Hillary

1.88

TRUMP

1.83

DonaldTrump

1.78

Bernie

1.66

trump

1.63

realDonaldTrump

1.59

Activations Density 0.016%

mentions of the name "Donald Trump"

mentions of Donald Trump

No Comments

No Known Activations