INDEX

Explanations

phrases related to change or transformation

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

enegger

-0.74

merce

-0.69

dale

-0.68

 Annotations

-0.68

gress

-0.68

 confir

-0.67

cffff

-0.66

MER

-0.65

 separates

-0.63

erness

-0.63

POSITIVE LOGITS

 tide

1.30

 tables

1.22

 tides

1.05

 knob

1.02

 Tables

1.01

 corner

1.00

 screws

0.95

 pages

0.95

 page

0.95

 spotlight

0.89

Activations Density 0.063%

phrases related to change or transformation

No Comments

No Known Activations