INDEX

Explanations

mentions of hotels

oai_token-act-pair · gpt-3.5-turbo

references to hotels and motels

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

owder

-0.80

alez

-0.75

xit

-0.75

 Anarchy

-0.73

 Matter

-0.71

erenn

-0.70

 Izan

-0.67

rd

-0.67

dash

-0.66

chie

-0.66

POSITIVE LOGITS

 hotel

1.14

 hotels

1.12

 Hotel

1.02

 accommodations

0.99

 accommodation

0.94

 guests

0.91

 motel

0.90

 resorts

0.86

Inn

0.84

 rooms

0.84

Activations Density 0.012%

mentions of hotels

references to hotels and motels

No Comments

No Known Activations