INDEX

Explanations

references to software vulnerabilities and their exploitation

oai_token-act-pair · gpt-3.5-turbo

terms related to exploitation, particularly in the context of vulnerabilities and unethical practices

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

ucket

-0.79

thood

-0.79

kos

-0.68

arant

-0.66

cone

-0.66

arat

-0.65

 mint

-0.64

ŃĶ

-0.63

cas

-0.62

HCR

-0.62

POSITIVE LOGITS

 loopholes

1.01

 weaknesses

1.00

 vulnerabilities

0.93

 exploited

0.91

CVE

0.88

ulner

0.87

 exploit

0.82

ulnerability

0.81

 exploiting

0.81

 exploits

0.81

Activations Density 0.058%

references to software vulnerabilities and their exploitation

terms related to exploitation, particularly in the context of vulnerabilities and unethical practices

No Comments

No Known Activations