INDEX

Explanations

words related to technology and software

oai_token-act-pair · gpt-3.5-turbo

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Rails

-0.65

 Boards

-0.64

USER

-0.62

blood

-0.61

CLR

-0.61

ACP

-0.58

 Weeks

-0.58

CHR

-0.58

 Learned

-0.58

 Mills

-0.56

POSITIVE LOGITS

aci

1.35

ement

1.35

Ã©s

1.34

ados

1.29

acion

1.24

iere

1.19

ament

1.19

ado

1.18

Ã©e

1.17

ient

1.17

Activations Density 0.161%

words related to technology and software

No Comments

No Known Activations