INDEX

Explanations

phrases indicating hypothetical situations or missed opportunities

oai_token-act-pair · gpt-3.5-turbo

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

/>

-0.60

 Definitions

-0.58

 Georgian

-0.54

 Americ

-0.53

 Spectre

-0.53

è¦ļéĨĴ

-0.53

 Outer

-0.53

 Outbreak

-0.52

 Intern

-0.51

 violet

-0.50

POSITIVE LOGITS

've

1.59

have

1.44

 have

1.42

 HAVE

1.18

 Have

1.16

Have

1.09

ered

0.91

been

0.91

Had

0.88

 haven

0.86

Activations Density 0.276%

phrases indicating hypothetical situations or missed opportunities

No Comments

No Known Activations