INDEX

Explanations

phrases related to vulnerabilities or weaknesses

oai_token-act-pair · gpt-3.5-turbo

references to vulnerability in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

zos

-0.72

 Collider

-0.71

inth

-0.68

atorium

-0.68

seller

-0.65

fare

-0.65

Ħ¢

-0.65

row

-0.64

hire

-0.63

rows

-0.63

POSITIVE LOGITS

 vulnerable

1.06

 vulner

0.89

ulnerable

0.87

 vulnerability

0.86

 vulnerabilities

0.86

ulnerability

0.79

ansas

0.76

 weaknesses

0.74

 exploited

0.70

 susceptible

0.67

Activations Density 0.023%

phrases related to vulnerabilities or weaknesses

references to vulnerability in various contexts

No Comments

No Known Activations