INDEX

Explanations

words related to sedation

oai_token-act-pair · gpt-3.5-turbo

variations of the term "seduce" and related terms associated with seduction

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

 Origin

-0.74

ball

-0.68

pei

-0.68

Hop

-0.65

foundation

-0.64

LAT

-0.63

 Organisation

-0.63

Aren

-0.63

Universal

-0.63

SHIP

-0.63

POSITIVE LOGITS

iments

1.24

uctive

1.20

entary

1.16

uct

1.07

uces

1.01

sed

0.99

uction

0.97

atives

0.95

ucing

0.95

icates

0.91

Activations Density 0.010%

words related to sedation

variations of the term "seduce" and related terms associated with seduction

No Comments

No Known Activations