INDEX

Explanations

personal pronouns indicating a strong personal stance or belief

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

catentry

-0.82

 Globe

-0.71

 Ammunition

-0.68

 Hunting

-0.66

elsen

-0.64

 Sund

-0.64

 Torrent

-0.64

antine

-0.60

 backdrop

-0.60

 Sever

-0.59

POSITIVE LOGITS

'll

1.31

're

1.24

've

1.17

'd

1.15

'm

1.07

 belong

1.01

owe

1.00

 shouldn

0.98

 wouldn

0.90

 HAVE

0.89

Activations Density 0.230%

personal pronouns indicating a strong personal stance or belief

No Known Activations