INDEX

Explanations

mentions or references to death

oai_token-act-pair · gpt-3.5-turbo

references to death

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

properties

-0.75

apan

-0.71

LET

-0.70

Ã±a

-0.68

ibe

-0.66

sb

-0.66

oline

-0.64

lett

-0.64

ï¸ı

-0.63

 vouchers

-0.63

POSITIVE LOGITS

 dead

3.84

dead

2.68

 Dead

2.41

 deceased

2.30

 DEAD

2.29

Dead

2.11

 lifeless

1.69

 corpses

1.56

 corpse

1.55

 dying

1.52

Activations Density 0.019%

mentions or references to death

references to death

No Comments

No Known Activations