INDEX

Explanations

references to hotels

oai_token-act-pair · gpt-3.5-turbo

mentions of hotels and lodging

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

ÏĢ

-0.71

udos

-0.70

gon

-0.68

clips

-0.67

 Fathers

-0.67

 reckoned

-0.65

WAR

-0.65

derived

-0.65

 DISTR

-0.65

 Democr

-0.64

POSITIVE LOGITS

 hotel

3.50

 hotels

2.85

 Hotel

2.57

 motel

2.44

 Marriott

2.00

 lodging

1.74

 casino

1.62

 hospitality

1.60

 Hilton

1.58

 airport

1.58

Activations Density 0.021%

references to hotels

mentions of hotels and lodging

No Comments

No Known Activations