INDEX

Explanations

phrases related to interpersonal interactions or events

oai_token-act-pair · gpt-3.5-turbo

instances of emotional statements or expressions related to personal experiences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

solete

-0.73

 rall

-0.71

 Coliseum

-0.67

udo

-0.64

ÃĥÃĤ

-0.64

espie

-0.64

actory

-0.62

PDATE

-0.61

undo

-0.61

raltar

-0.61

POSITIVE LOGITS

 herself

1.88

 husband

1.20

 boyfriend

1.07

 breasts

1.05

 pregnant

1.02

 maternity

1.00

 breastfeeding

0.99

 maid

0.99

 childbirth

0.97

 pregnancy

0.96

Activations Density 0.980%

phrases related to interpersonal interactions or events

instances of emotional statements or expressions related to personal experiences

No Comments

No Known Activations