INDEX

Explanations

names of people or entities ending in 'y'

oai_token-act-pair · gpt-3.5-turbo

references to people and specific entities

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

CAP

-0.85

 Manit

-0.84

ãĤ¬

-0.81

 Citadel

-0.80

 Feder

-0.77

 Covenant

-0.72

 conduc

-0.72

 ISIL

-0.72

 arrang

-0.71

 appendix

-0.71

POSITIVE LOGITS

1.51

1.47

ys

1.36

yon

1.35

yg

1.33

yu

1.29

ya

1.27

yy

1.26

yan

1.24

YS

1.24

Activations Density 0.322%

names of people or entities ending in 'y'

references to people and specific entities

No Comments

No Known Activations