INDEX

Explanations

phrases related to legal or political actions

oai_token-act-pair · gpt-3.5-turbo

items related to significant actions or events

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

 dispers

-0.71

 Belg

-0.71

Jem

-0.68

 harness

-0.68

 assemb

-0.68

 range

-0.68

 mamm

-0.66

 mate

-0.65

 handshake

-0.65

 embark

-0.65

POSITIVE LOGITS

ª

1.44

¡

1.37

¹

1.30

¬

1.29

Ń

1.28

«

1.27

Į

1.26

Ŀ

1.25

ĺ

1.24

²

1.23

Activations Density 0.114%

phrases related to legal or political actions

items related to significant actions or events

No Comments

No Known Activations