INDEX

Explanations

technical terms related to cyber attacks and security vulnerabilities

oai_token-act-pair · gpt-3.5-turbo

references to cybersecurity threats and attacks

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

ikk

-0.86

essional

-0.82

thumbnails

-0.75

iatrics

-0.75

 Colon

-0.74

zl

-0.73

âĢ¢âĢ¢

-0.73

rians

-0.71

 Yards

-0.71

Beer

-0.70

POSITIVE LOGITS

CVE

1.22

 malware

1.21

CVE

1.20

 hackers

1.05

 vulnerabilities

1.04

 exploit

1.03

 vulnerability

1.03

 ransomware

1.00

 cryptographic

0.99

 hacker

0.99

Activations Density 0.187%

technical terms related to cyber attacks and security vulnerabilities

references to cybersecurity threats and attacks

No Comments

No Known Activations