INDEX

Explanations

derogatory terms or insults directed towards individuals

oai_token-act-pair · gpt-3.5-turbo

derogatory terms and insults aimed at individuals or groups

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

 conclud

-0.72

 Printed

-0.71

 survives

-0.71

 Surviv

-0.70

foreseen

-0.68

ortality

-0.67

 traject

-0.67

 ORDER

-0.66

igsaw

-0.64

 longitudinal

-0.63

POSITIVE LOGITS

 liar

1.03

 coward

0.93

 irresponsible

0.92

 traitor

0.91

 unfit

0.90

 hypocr

0.89

 unworthy

0.87

 disgrace

0.84

 disrespectful

0.82

 insensitive

0.81

Activations Density 0.387%

derogatory terms or insults directed towards individuals

derogatory terms and insults aimed at individuals or groups

No Comments

No Known Activations