INDEX

Explanations

keywords related to cybersecurity exploits

oai_token-act-pair · gpt-3.5-turbo

terms related to exploitation and vulnerabilities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

arat

-0.78

ael

-0.68

prints

-0.66

hair

-0.65

 Presbyter

-0.65

 captcha

-0.64

present

-0.63

quartered

-0.62

é¾įåĸļå£«

-0.62

thood

-0.61

POSITIVE LOGITS

oit

1.08

 exploit

0.96

 loopholes

0.94

CVE

0.92

 exploiting

0.92

 weaknesses

0.91

ulner

0.88

 exploited

0.87

ulnerability

0.84

ãĥ¼ãĥĨãĤ£

0.80

Activations Density 0.012%

keywords related to cybersecurity exploits

terms related to exploitation and vulnerabilities

No Comments

No Known Activations