INDEX

Explanations

references to interactions with law enforcement or police involvement

oai_token-act-pair · gpt-3.5-turbo

references to police involvement in various situations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

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

yip

-0.85

bernatorial

-0.81

ocene

-0.78

cast

-0.70

apest

-0.70

opian

-0.69

ãĤ¦

-0.68

edia

-0.68

resy

-0.65

 Chamber

-0.64

POSITIVE LOGITS

 officers

1.25

 detectives

1.09

 negotiators

1.03

 investigators

1.00

 departments

0.93

Officers

0.92

 scanners

0.91

 Officers

0.90

 snipers

0.86

 officer

0.86

Activations Density 0.066%

references to interactions with law enforcement or police involvement

references to police involvement in various situations

No Comments

No Known Activations