INDEX

Explanations

references to physical objects or entities located on a person's body

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

 ÂŃ

-0.76

âĢ¢âĢ¢

-0.75

igible

-0.66

quished

-0.65

fitted

-0.64

ocused

-0.61

ãĤ¡

-0.61

ogged

-0.60

Said

-0.60

rontal

-0.60

POSITIVE LOGITS

 doesnt

1.55

 differs

1.40

 seems

1.30

 consists

1.28

 tends

1.23

 does

1.23

 lacks

1.21

 resembles

1.20

 behaves

1.18

is

1.17

Activations Density 0.753%

references to physical objects or entities located on a person's body

No Comments

No Known Activations