INDEX

Explanations

instances of the word "automatic"

oai_token-act-pair · gpt-3.5-turbo

terms associated with automatic mechanisms or systems

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 McKenna

-0.83

rug

-0.80

borough

-0.79

our

-0.78

hoff

-0.77

adr

-0.76

eters

-0.76

ors

-0.76

ALE

-0.74

or

-0.73

POSITIVE LOGITS

 automatic

1.20

 confir

1.09

 captcha

1.06

 transmission

1.02

omatic

1.02

 override

1.01

 induction

0.99

 exha

0.96

 derog

0.93

automatic

0.92

Activations Density 0.008%

instances of the word "automatic"

terms associated with automatic mechanisms or systems

No Comments

No Known Activations