INDEX

Explanations

mentions of cars

oai_token-act-pair · gpt-3.5-turbo

references to cars and automobiles

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

wu

-0.81

unity

-0.66

ledged

-0.66

 Primordial

-0.66

ionage

-0.66

Ëľ

-0.65

åħī

-0.64

parency

-0.62

 notwithstanding

-0.62

NESS

-0.62

POSITIVE LOGITS

car

3.38

 cars

2.49

 automobile

2.28

 vehicle

2.19

car

2.15

 Cars

1.90

 sedan

1.85

 automobiles

1.78

SUV

1.77

cars

1.75

Activations Density 0.023%

mentions of cars

references to cars and automobiles

No Comments

No Known Activations