INDEX

Explanations

specific mentions of vulnerabilities in the text

oai_token-act-pair · gpt-3.5-turbo

terms related to security vulnerabilities

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

icle

-1.00

icles

-0.99

zos

-0.88

kos

-0.85

omers

-0.78

piece

-0.77

omer

-0.75

İ

-0.74

agher

-0.74

axis

-0.73

POSITIVE LOGITS

CVE

1.14

ulnerability

1.03

ulner

1.03

CVE

0.99

 vulner

0.99

 vulnerabilities

0.98

 vulnerability

0.93

 flaws

0.84

 weaknesses

0.82

 patched

0.81

Activations Density 0.019%

specific mentions of vulnerabilities in the text

terms related to security vulnerabilities

No Comments

No Known Activations