INDEX

Explanations

phrases related to locations or directions

oai_token-act-pair · gpt-3.5-turbo

descriptors of intensity or physical attributes such as depth, height, and speed

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

 Divinity

-0.71

ITNESS

-0.69

CoC

-0.69

vous

-0.66

âķĲâķĲ

-0.63

Quantity

-0.61

âĢ¢âĢ¢âĢ¢âĢ¢

-0.60

cend

-0.59

arth

-0.59

 Offic

-0.55

POSITIVE LOGITS

aneously

1.23

 enough

1.20

handedly

0.99

heartedly

0.88

 distances

0.86

enough

0.86

iously

0.83

itud

0.83

eners

0.82

ened

0.78

Activations Density 0.256%

phrases related to locations or directions

descriptors of intensity or physical attributes such as depth, height, and speed

No Comments

No Known Activations