INDEX

Explanations

proper names related to individuals

oai_token-act-pair · gpt-3.5-turbo

names of individuals and their associated attributes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

alysed

-0.74

ĻĤ

-0.70

quartered

-0.68

Asian

-0.67

VALUE

-0.67

 indoor

-0.66

incial

-0.65

icultural

-0.65

vana

-0.65

ixty

-0.65

POSITIVE LOGITS

McH

1.31

 McCarthy

1.26

 Daniels

1.25

 Byrd

1.24

Poe

1.23

McN

1.23

 Lowry

1.22

 Kling

1.22

 Faul

1.21

 Collins

1.21

Activations Density 0.212%

proper names related to individuals

names of individuals and their associated attributes

No Comments

No Known Activations