INDEX

Explanations

phrases related to paying attention or being observant

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

igsaw

-0.71

è¦ļéĨĴ

-0.70

execute

-0.63

------

-0.63

swick

-0.62

 Species

-0.61

ãĤ¬

-0.61

ãĤ¼ãĤ¦ãĤ¹

-0.61

yrinth

-0.61

lihood

-0.60

POSITIVE LOGITS

 attention

1.43

 homage

1.34

 tribute

1.24

 heed

1.23

 dearly

1.17

lip

1.11

 Attention

1.09

 dues

1.07

 respects

1.04

 dividends

0.98

Activations Density 0.047%

phrases related to paying attention or being observant

No Comments

No Known Activations