INDEX

Explanations

the name "Irvin" with varying levels of emphasis

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

ãģ®éŃĶ

-0.67

 Rampage

-0.64

 compliment

-0.64

ezvous

-0.63

 Archdemon

-0.63

 Hots

-0.62

eries

-0.62

Ou

-0.61

 CLSID

-0.58

 messenger

-0.58

POSITIVE LOGITS

rational

1.29

regular

1.24

respective

1.20

resist

1.19

relevant

1.15

responsible

1.13

idium

1.07

anian

1.00

rig

0.99

replace

0.95

Activations Density 0.021%

the name "Irvin" with varying levels of emphasis

No Comments

No Known Activations