INDEX

Explanations

derogatory and offensive language

oai_token-act-pair · gpt-3.5-turbo

derogatory terms and insults

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

tnc

-0.78

earchers

-0.76

Decre

-0.75

paren

-0.74

Recomm

-0.73

restricted

-0.70

soType

-0.69

aeda

-0.69

catentry

-0.69

encers

-0.69

POSITIVE LOGITS

fuck

1.10

 bitch

1.01

 bastard

1.01

hole

0.99

 idiots

0.98

holes

0.98

 asses

0.97

bags

0.96

gery

0.96

 Bastard

0.94

Activations Density 0.168%

derogatory and offensive language

derogatory terms and insults

No Comments

No Known Activations