INDEX

Explanations

mentions of violations or adherence to international laws

oai_token-act-pair · gpt-3.5-turbo

references to international law and its violations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

 Tate

-0.76

 Volunte

-0.67

 Bomber

-0.63

ï¸ı

-0.62

Ò

-0.61

CV

-0.61

 flyer

-0.60

 throats

-0.58

 Sold

-0.58

 Horde

-0.58

POSITIVE LOGITS

 enforcement

1.26

enforcement

1.01

suit

0.96

suits

0.92

 Enforcement

0.91

 abiding

0.88

lessness

0.85

 governing

0.83

makers

0.81

 forb

0.80

Activations Density 0.056%

mentions of violations or adherence to international laws

references to international law and its violations

No Comments

No Known Activations