INDEX

Explanations

mentions or occurrences of police involvement in incidents

oai_token-act-pair · gpt-3.5-turbo

mentions of police or law enforcement activities

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

bernatorial

-0.90

ocene

-0.73

ggles

-0.64

osi

-0.64

pub

-0.61

 Chamber

-0.61

 vetoed

-0.59

yip

-0.59

gets

-0.59

Wo

-0.59

POSITIVE LOGITS

 officers

1.04

 investigators

1.00

 detectives

0.98

 investigating

0.89

 Detect

0.85

Officers

0.85

 Constable

0.83

 Officers

0.83

 Investigators

0.82

 detective

0.81

Activations Density 0.076%

mentions or occurrences of police involvement in incidents

mentions of police or law enforcement activities

No Comments

No Known Activations