INDEX

Explanations

details related to crime incidents

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

references to criminal activities or incidents

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 12-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.hook_resid_post

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_post

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

 balance

-0.71

 stagnation

-0.69

 greatness

-0.64

 univers

-0.63

 balanced

-0.63

 longitudinal

-0.62

SPL

-0.62

 parity

-0.60

 stagn

-0.60

 allev

-0.59

POSITIVE LOGITS

 gunmen

1.18

 robberies

1.07

 robbers

1.06

 explosives

1.05

 assailants

1.04

nikov

1.02

 burgl

1.01

Police

1.01

 gunman

0.98

Officers

0.98

Activations Density 0.635%

details related to crime incidents

references to criminal activities or incidents

No Comments

No Known Activations