INDEX

Explanations

items or actions related to physical objects or plans

oai_token-act-pair · gpt-3.5-turbo

references to lists or items being enumerated

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

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Contract

-0.68

eye

-0.64

 Chow

-0.64

BO

-0.63

cap

-0.62

God

-0.62

 call

-0.62

 light

-0.62

 Card

-0.61

 Dust

-0.60

POSITIVE LOGITS

ĸļå£«

1.23

EStream

1.12

¥ŀ

1.02

DeliveryDate

1.00

Orderable

0.99

ÃĥÃĤÃĥÃĤ

0.98

.):

0.94

ailability

0.93

ò

0.92

ñ

0.90

Activations Density 0.245%

items or actions related to physical objects or plans

references to lists or items being enumerated

No Comments

No Known Activations