INDEX

Explanations

criminal activities or behaviors

oai_token-act-pair · gpt-3.5-turbo

phrases related to criminal activity and charges

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

isSpecialOrderable

-0.79

sson

-0.72

ishops

-0.70

ciation

-0.69

fortunately

-0.69

thank

-0.67

DragonMagazine

-0.67

 lineback

-0.66

gradation

-0.66

reci

-0.65

POSITIVE LOGITS

 illegally

1.25

 illegal

1.24

 unlawfully

1.21

 improper

1.21

 unauthorized

1.19

 indecent

1.17

 improperly

1.15

 stolen

1.14

 unlawful

1.13

 illicit

1.13

Activations Density 0.281%

criminal activities or behaviors

phrases related to criminal activity and charges

No Comments

No Known Activations