INDEX

Explanations

terms related to security and vulnerability

oai_token-act-pair · gpt-3.5-turbo

terms related to vulnerability and security issues

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

ggles

-0.69

CVE

-0.68

FK

-0.68

 Tarant

-0.67

 abbre

-0.63

UDP

-0.60

 Hydra

-0.57

 clarification

-0.57

 Pompe

-0.57

 Lethal

-0.57

POSITIVE LOGITS

 themselves

1.26

selves

1.18

 selves

1.02

 academ

0.99

their

0.99

 psychologically

0.97

 financially

0.97

 their

0.96

 THEIR

0.95

Their

0.91

Activations Density 0.559%

terms related to security and vulnerability

terms related to vulnerability and security issues

No Known Activations