INDEX

Explanations

references to the former US Attorney General Jeff Sessions

oai_token-act-pair · gpt-3.5-turbo

mentions of Jeff Sessions

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

VID

-0.73

 pity

-0.68

product

-0.68

oard

-0.68

 Destination

-0.67

iHUD

-0.66

 Trib

-0.65

 Mortal

-0.65

OPS

-0.63

 Squid

-0.62

POSITIVE LOGITS

 Sessions

1.11

essions

0.86

achus

0.86

alty

0.86

nyder

0.85

esson

0.84

agne

0.82

bourg

0.79

adiq

0.78

oka

0.76

Activations Density 0.013%

references to the former US Attorney General Jeff Sessions

mentions of Jeff Sessions

No Comments

No Known Activations