INDEX

Explanations

phrases indicating items or people being part of a group or set

oai_token-act-pair · gpt-3.5-turbo

phrases that highlight the concept of "togetherness" or "association."

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

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

Not in Any Lists

No Comments

Negative Logits

orius

-0.81

BACK

-0.75

imester

-0.71

chin

-0.71

ager

-0.70

iste

-0.70

Leaks

-0.70

igh

-0.69

ade

-0.68

auri

-0.68

POSITIVE LOGITS

 others

1.18

 other

1.14

 countless

1.00

 fellow

0.90

 assorted

0.88

 innumerable

0.88

 dozens

0.87

 numerous

0.83

 some

0.83

 plenty

0.82

Activations Density 0.088%

phrases indicating items or people being part of a group or set

phrases that highlight the concept of "togetherness" or "association."

No Comments

No Known Activations