INDEX

Explanations

organizations and professional titles with an emphasis on manufacturing and trade

oai_token-act-pair · gpt-3.5-turbo

words related to various professional associations and organizations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

lier

-0.75

efully

-0.72

mun

-0.70

liest

-0.70

ctors

-0.66

times

-0.65

lying

-0.65

ris

-0.63

bed

-0.63

 condition

-0.61

POSITIVE LOGITS

hip

1.22

paces

1.18

mith

1.15

hift

1.11

aurus

1.08

pace

1.03

hips

1.02

Ltd

1.00

 Association

1.00

ystem

0.99

Activations Density 0.158%

organizations and professional titles with an emphasis on manufacturing and trade

words related to various professional associations and organizations

No Comments

No Known Activations