INDEX

Explanations

phrases related to criminal investigations and law enforcement activities

oai_token-act-pair · gpt-3.5-turbo

terms related to criminal investigations and hate crimes

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Kiw

-0.90

ussie

-0.89

 Hampton

-0.78

 Tyson

-0.78

AFL

-0.78

 Fallon

-0.75

 Berkshire

-0.74

 Winston

-0.74

 Lincoln

-0.71

 Hughes

-0.71

POSITIVE LOGITS

Ä±

1.22

 Azerbai

1.21

ÅŁ

1.12

 Municip

1.07

ÄŁ

1.06

Ä

1.05

 testimonies

1.04

 municipality

1.00

ãĢĲ

0.97

ãĢİ

0.96

Activations Density 1.120%

phrases related to criminal investigations and law enforcement activities

terms related to criminal investigations and hate crimes

No Comments

No Known Activations