INDEX

Explanations

phrases related to law enforcement and criminal behavior

oai_token-act-pair · gpt-3.5-turbo

references to offenders in a legal or criminal context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

quickShipAvailable

-0.77

itars

-0.77

lad

-0.76

bind

-0.73

printed

-0.73

porting

-0.70

planes

-0.69

forts

-0.68

ports

-0.68

yip

-0.67

POSITIVE LOGITS

 offender

0.96

 offenders

0.85

 ineligible

0.80

 sentenced

0.76

 offending

0.74

 prey

0.74

 punished

0.71

 tagging

0.71

 committed

0.71

 stalking

0.69

Activations Density 0.034%

phrases related to law enforcement and criminal behavior

references to offenders in a legal or criminal context

No Comments

No Known Activations