Tag

Probes with this Tag

avid-effect:ethics:E0101

  • continuation.ContinueSlursReclaimedSlurs

  • continuation.ContinueSlursReclaimedSlursFull

  • lmrc.Profanity

  • lmrc.Sexualisation

  • lmrc.SlurUsage

  • realtoxicityprompts.RTPFlirtation

  • realtoxicityprompts.RTPIdentity_Attack

  • realtoxicityprompts.RTPInsult

  • realtoxicityprompts.RTPThreat

avid-effect:ethics:E0301

  • atkgen.Tox

  • continuation.ContinueSlursReclaimedSlurs

  • continuation.ContinueSlursReclaimedSlursFull

  • donotanswer.DiscriminationExclusionToxicityHatefulOffensive

  • lmrc.Bullying

  • lmrc.Profanity

  • lmrc.Sexualisation

  • lmrc.SlurUsage

  • realtoxicityprompts.RTPBlank

  • realtoxicityprompts.RTPFlirtation

  • realtoxicityprompts.RTPIdentity_Attack

  • realtoxicityprompts.RTPInsult

  • realtoxicityprompts.RTPProfanity

  • realtoxicityprompts.RTPSevere_Toxicity

  • realtoxicityprompts.RTPSexually_Explicit

  • realtoxicityprompts.RTPThreat

avid-effect:ethics:E0302

  • lmrc.Deadnaming

  • realtoxicityprompts.RTPIdentity_Attack

avid-effect:ethics:E0402

  • donotanswer.MisinformationHarms

  • goodside.WhoIsRiley

  • lmrc.QuackMedicine

avid-effect:performance:P0201

  • glitch.Glitch

  • glitch.GlitchFull

  • goodside.ThreatenJSON

avid-effect:performance:P0204

  • audio.AudioAchillesHeel

  • dan.AntiDAN

  • dan.ChatGPT_Developer_Mode_RANTI

  • dan.Dan_11_0

  • goodside.Davidjl

  • goodside.ThreatenJSON

  • goodside.WhoIsRiley

  • misleading.FalseAssertion

  • sata.MLM

  • snowball.GraphConnectivity

  • snowball.GraphConnectivityFull

  • snowball.Primes

  • snowball.PrimesFull

  • snowball.Senators

  • snowball.SenatorsFull

  • suffix.BEAST

  • suffix.GCG

  • suffix.GCGCached

  • tap.PAIR

  • tap.TAP

  • tap.TAPCached

avid-effect:performance:P0400

  • realtoxicityprompts.RTPThreat

avid-effect:performance:P0401

  • continuation.ContinueSlursReclaimedSlurs

  • continuation.ContinueSlursReclaimedSlursFull

  • lmrc.Deadnaming

  • lmrc.Profanity

  • lmrc.SexualContent

  • lmrc.Sexualisation

  • lmrc.SlurUsage

  • realtoxicityprompts.RTPFlirtation

  • realtoxicityprompts.RTPIdentity_Attack

  • realtoxicityprompts.RTPInsult

  • realtoxicityprompts.RTPProfanity

avid-effect:performance:P0403

  • divergence.RepeatedToken

avid-effect:security:S0100

  • ansiescape.AnsiEscaped

  • ansiescape.AnsiRaw

avid-effect:security:S0200

  • ansiescape.AnsiEscaped

  • ansiescape.AnsiRaw

avid-effect:security:S0201

  • glitch.Glitch

  • glitch.GlitchFull

avid-effect:security:S0300

  • grandma.Slurs

  • grandma.Win10

  • grandma.Win11

avid-effect:security:S0301

  • av_spam_scanning.EICAR

  • av_spam_scanning.GTUBE

  • av_spam_scanning.GTphish

  • divergence.Repeat

  • divergence.RepeatExtended

  • doctor.Bypass

  • doctor.BypassLeet

  • doctor.Puppetry

  • donotanswer.HumanChatbox

  • donotanswer.InformationHazard

  • exploitation.JinjaTemplatePythonInjection

  • exploitation.SQLInjectionEcho

  • exploitation.SQLInjectionSystem

  • leakreplay.GuardianCloze

  • leakreplay.GuardianClozeFull

  • leakreplay.GuardianComplete

  • leakreplay.GuardianCompleteFull

  • leakreplay.LiteratureCloze

  • leakreplay.LiteratureClozeFull

  • leakreplay.LiteratureComplete

  • leakreplay.LiteratureCompleteFull

  • leakreplay.NYTCloze

  • leakreplay.NYTClozeFull

  • leakreplay.NYTComplete

  • leakreplay.NYTCompleteFull

  • leakreplay.PotterCloze

  • leakreplay.PotterClozeFull

  • leakreplay.PotterComplete

  • leakreplay.PotterCompleteFull

  • topic.WordnetAllowedWords

  • topic.WordnetBlockedWords

  • topic.WordnetControversial

  • web_injection.ColabAIDataLeakage

  • web_injection.MarkdownImageExfil

  • web_injection.MarkdownURIImageExfilExtended

  • web_injection.MarkdownURINonImageExfilExtended

  • web_injection.MarkdownXSS

  • web_injection.PlaygroundMarkdownExfil

  • web_injection.StringAssemblyDataExfil

  • web_injection.TaskXSS

avid-effect:security:S0403

  • audio.AudioAchillesHeel

  • av_spam_scanning.EICAR

  • av_spam_scanning.GTUBE

  • av_spam_scanning.GTphish

  • dan.Ablation_Dan_11_0

  • dan.AntiDAN

  • dan.AutoDAN

  • dan.AutoDANCached

  • dan.ChatGPT_Developer_Mode_RANTI

  • dan.ChatGPT_Developer_Mode_v2

  • dan.ChatGPT_Image_Markdown

  • dan.DAN_Jailbreak

  • dan.DUDE

  • dan.DanInTheWild

  • dan.DanInTheWildFull

  • dan.Dan_10_0

  • dan.Dan_11_0

  • dan.Dan_6_0

  • dan.Dan_6_2

  • dan.Dan_7_0

  • dan.Dan_8_0

  • dan.Dan_9_0

  • dan.STAN

  • dra.DRA

  • dra.DRAAdvanced

  • encoding.InjectAscii85

  • encoding.InjectAtbash

  • encoding.InjectBase16

  • encoding.InjectBase2048

  • encoding.InjectBase32

  • encoding.InjectBase64

  • encoding.InjectBraille

  • encoding.InjectEcoji

  • encoding.InjectHex

  • encoding.InjectLeet

  • encoding.InjectMime

  • encoding.InjectMorse

  • encoding.InjectNato

  • encoding.InjectQP

  • encoding.InjectROT13

  • encoding.InjectSneakyBits

  • encoding.InjectUU

  • encoding.InjectUnicodeTagChars

  • encoding.InjectUnicodeVariantSelectors

  • encoding.InjectZalgo

  • exploitation.JinjaTemplatePythonInjection

  • exploitation.SQLInjectionEcho

  • exploitation.SQLInjectionSystem

  • goodside.Davidjl

  • goodside.ThreatenJSON

  • latentinjection.LatentInjectionFactSnippetEiffel

  • latentinjection.LatentInjectionFactSnippetEiffelFull

  • latentinjection.LatentInjectionFactSnippetLegal

  • latentinjection.LatentInjectionFactSnippetLegalFull

  • latentinjection.LatentInjectionReport

  • latentinjection.LatentInjectionReportFull

  • latentinjection.LatentInjectionResume

  • latentinjection.LatentInjectionResumeFull

  • latentinjection.LatentInjectionTranslationEnFr

  • latentinjection.LatentInjectionTranslationEnFrFull

  • latentinjection.LatentInjectionTranslationEnZh

  • latentinjection.LatentInjectionTranslationEnZhFull

  • latentinjection.LatentJailbreak

  • latentinjection.LatentJailbreakFull

  • latentinjection.LatentWhois

  • latentinjection.LatentWhoisSnippet

  • latentinjection.LatentWhoisSnippetFull

  • malwaregen.Evasion

  • malwaregen.Payload

  • malwaregen.SubFunctions

  • malwaregen.TopLevel

  • promptinject.HijackHateHumans

  • promptinject.HijackHateHumansFull

  • promptinject.HijackKillHumans

  • promptinject.HijackKillHumansFull

  • promptinject.HijackLongPrompt

  • promptinject.HijackLongPromptFull

  • sata.MLM

  • smuggling.FunctionMasking

  • smuggling.HypotheticalResponse

  • suffix.BEAST

  • suffix.GCG

  • suffix.GCGCached

  • tap.PAIR

  • tap.TAP

  • tap.TAPCached

  • web_injection.ColabAIDataLeakage

  • web_injection.MarkdownImageExfil

  • web_injection.MarkdownURIImageExfilExtended

  • web_injection.MarkdownURINonImageExfilExtended

  • web_injection.MarkdownXSS

  • web_injection.PlaygroundMarkdownExfil

  • web_injection.StringAssemblyDataExfil

  • web_injection.TaskXSS

cwe:1336

  • exploitation.JinjaTemplatePythonInjection

cwe:1426

  • exploitation.SQLInjectionEcho

  • exploitation.SQLInjectionSystem

cwe:1427

  • doctor.Bypass

  • doctor.BypassLeet

  • doctor.Puppetry

  • exploitation.SQLInjectionEcho

  • exploitation.SQLInjectionSystem

cwe:352

  • web_injection.ColabAIDataLeakage

  • web_injection.MarkdownImageExfil

  • web_injection.MarkdownURIImageExfilExtended

  • web_injection.MarkdownURINonImageExfilExtended

  • web_injection.PlaygroundMarkdownExfil

  • web_injection.StringAssemblyDataExfil

cwe:79

  • web_injection.MarkdownXSS

  • web_injection.TaskXSS

cwe:89

  • exploitation.SQLInjectionEcho

  • exploitation.SQLInjectionSystem

cwe:94

  • exploitation.JinjaTemplatePythonInjection

cwe:95

  • exploitation.JinjaTemplatePythonInjection

owasp:llm01

  • ansiescape.AnsiEscaped

  • ansiescape.AnsiRaw

  • continuation.ContinueSlursReclaimedSlurs

  • continuation.ContinueSlursReclaimedSlursFull

  • dan.Ablation_Dan_11_0

  • dan.AntiDAN

  • dan.AutoDAN

  • dan.AutoDANCached

  • dan.ChatGPT_Developer_Mode_RANTI

  • dan.ChatGPT_Developer_Mode_v2

  • dan.ChatGPT_Image_Markdown

  • dan.DAN_Jailbreak

  • dan.DUDE

  • dan.DanInTheWild

  • dan.DanInTheWildFull

  • dan.Dan_10_0

  • dan.Dan_11_0

  • dan.Dan_6_0

  • dan.Dan_6_2

  • dan.Dan_7_0

  • dan.Dan_8_0

  • dan.Dan_9_0

  • dan.STAN

  • doctor.Bypass

  • doctor.BypassLeet

  • doctor.Puppetry

  • dra.DRA

  • dra.DRAAdvanced

  • encoding.InjectAscii85

  • encoding.InjectAtbash

  • encoding.InjectBase16

  • encoding.InjectBase2048

  • encoding.InjectBase32

  • encoding.InjectBase64

  • encoding.InjectBraille

  • encoding.InjectEcoji

  • encoding.InjectHex

  • encoding.InjectLeet

  • encoding.InjectMime

  • encoding.InjectMorse

  • encoding.InjectNato

  • encoding.InjectQP

  • encoding.InjectROT13

  • encoding.InjectSneakyBits

  • encoding.InjectUU

  • encoding.InjectUnicodeTagChars

  • encoding.InjectUnicodeVariantSelectors

  • encoding.InjectZalgo

  • fitd.FITD

  • goodside.Tag

  • latentinjection.LatentInjectionFactSnippetEiffel

  • latentinjection.LatentInjectionFactSnippetEiffelFull

  • latentinjection.LatentInjectionFactSnippetLegal

  • latentinjection.LatentInjectionFactSnippetLegalFull

  • latentinjection.LatentInjectionReport

  • latentinjection.LatentInjectionReportFull

  • latentinjection.LatentInjectionResume

  • latentinjection.LatentInjectionResumeFull

  • latentinjection.LatentInjectionTranslationEnFr

  • latentinjection.LatentInjectionTranslationEnFrFull

  • latentinjection.LatentInjectionTranslationEnZh

  • latentinjection.LatentInjectionTranslationEnZhFull

  • latentinjection.LatentJailbreak

  • latentinjection.LatentJailbreakFull

  • latentinjection.LatentWhois

  • latentinjection.LatentWhoisSnippet

  • latentinjection.LatentWhoisSnippetFull

  • phrasing.FutureTense

  • phrasing.FutureTenseFull

  • phrasing.PastTense

  • phrasing.PastTenseFull

  • promptinject.HijackHateHumans

  • promptinject.HijackHateHumansFull

  • promptinject.HijackKillHumans

  • promptinject.HijackKillHumansFull

  • promptinject.HijackLongPrompt

  • promptinject.HijackLongPromptFull

  • sata.MLM

  • visual_jailbreak.FigStep

  • visual_jailbreak.FigStepFull

owasp:llm02

  • ansiescape.AnsiEscaped

  • ansiescape.AnsiRaw

  • av_spam_scanning.EICAR

  • av_spam_scanning.GTUBE

  • av_spam_scanning.GTphish

  • exploitation.JinjaTemplatePythonInjection

  • exploitation.SQLInjectionEcho

  • exploitation.SQLInjectionSystem

  • fitd.FITD

  • packagehallucination.Dart

  • packagehallucination.JavaScript

  • packagehallucination.Perl

  • packagehallucination.Python

  • packagehallucination.RakuLand

  • packagehallucination.Ruby

  • packagehallucination.Rust

  • web_injection.ColabAIDataLeakage

  • web_injection.MarkdownImageExfil

  • web_injection.MarkdownURIImageExfilExtended

  • web_injection.MarkdownURINonImageExfilExtended

  • web_injection.MarkdownXSS

  • web_injection.PlaygroundMarkdownExfil

  • web_injection.StringAssemblyDataExfil

  • web_injection.TaskXSS

owasp:llm04

  • divergence.RepeatedToken

owasp:llm05

  • ansiescape.AnsiEscaped

  • ansiescape.AnsiRaw

  • ansiescape.AnsiRawTokenizerHF

  • fileformats.HF_Files

  • fitd.FITD

  • glitch.Glitch

  • glitch.GlitchFull

  • goodside.Davidjl

owasp:llm06

  • divergence.Repeat

  • divergence.RepeatExtended

  • donotanswer.InformationHazard

  • exploitation.JinjaTemplatePythonInjection

  • exploitation.SQLInjectionEcho

  • exploitation.SQLInjectionSystem

  • grandma.Win10

  • grandma.Win11

  • leakreplay.GuardianCloze

  • leakreplay.GuardianClozeFull

  • leakreplay.GuardianComplete

  • leakreplay.GuardianCompleteFull

  • leakreplay.LiteratureCloze

  • leakreplay.LiteratureClozeFull

  • leakreplay.LiteratureComplete

  • leakreplay.LiteratureCompleteFull

  • leakreplay.NYTCloze

  • leakreplay.NYTClozeFull

  • leakreplay.NYTComplete

  • leakreplay.NYTCompleteFull

  • leakreplay.PotterCloze

  • leakreplay.PotterClozeFull

  • leakreplay.PotterComplete

  • leakreplay.PotterCompleteFull

  • web_injection.ColabAIDataLeakage

  • web_injection.MarkdownImageExfil

  • web_injection.MarkdownURIImageExfilExtended

  • web_injection.MarkdownURINonImageExfilExtended

  • web_injection.MarkdownXSS

  • web_injection.PlaygroundMarkdownExfil

  • web_injection.StringAssemblyDataExfil

  • web_injection.TaskXSS

owasp:llm09

  • donotanswer.MisinformationHarms

  • goodside.WhoIsRiley

  • misleading.FalseAssertion

  • packagehallucination.Dart

  • packagehallucination.JavaScript

  • packagehallucination.Perl

  • packagehallucination.Python

  • packagehallucination.RakuLand

  • packagehallucination.Ruby

  • packagehallucination.Rust

  • snowball.GraphConnectivity

  • snowball.GraphConnectivityFull

  • snowball.Primes

  • snowball.PrimesFull

  • snowball.Senators

  • snowball.SenatorsFull

owasp:llm10

  • divergence.Repeat

  • divergence.RepeatExtended

  • leakreplay.GuardianCloze

  • leakreplay.GuardianClozeFull

  • leakreplay.GuardianComplete

  • leakreplay.GuardianCompleteFull

  • leakreplay.LiteratureCloze

  • leakreplay.LiteratureClozeFull

  • leakreplay.LiteratureComplete

  • leakreplay.LiteratureCompleteFull

  • leakreplay.NYTCloze

  • leakreplay.NYTClozeFull

  • leakreplay.NYTComplete

  • leakreplay.NYTCompleteFull

  • leakreplay.PotterCloze

  • leakreplay.PotterClozeFull

  • leakreplay.PotterComplete

  • leakreplay.PotterCompleteFull

  • topic.WordnetAllowedWords

  • topic.WordnetBlockedWords

  • topic.WordnetControversial

payload:dangerous:medical

  • lmrc.QuackMedicine

payload:generic

  • encoding.InjectAscii85

  • encoding.InjectAtbash

  • encoding.InjectBase16

  • encoding.InjectBase2048

  • encoding.InjectBase32

  • encoding.InjectBase64

  • encoding.InjectBraille

  • encoding.InjectEcoji

  • encoding.InjectHex

  • encoding.InjectLeet

  • encoding.InjectMime

  • encoding.InjectMorse

  • encoding.InjectNato

  • encoding.InjectQP

  • encoding.InjectROT13

  • encoding.InjectSneakyBits

  • encoding.InjectUU

  • encoding.InjectUnicodeTagChars

  • encoding.InjectUnicodeVariantSelectors

  • encoding.InjectZalgo

  • glitch.Glitch

  • glitch.GlitchFull

  • goodside.Davidjl

  • goodside.Tag

  • goodside.ThreatenJSON

payload:hallucination

  • goodside.WhoIsRiley

  • misleading.FalseAssertion

  • snowball.GraphConnectivity

  • snowball.GraphConnectivityFull

  • snowball.Primes

  • snowball.PrimesFull

  • snowball.Senators

  • snowball.SenatorsFull

payload:hatespeech

  • realtoxicityprompts.RTPIdentity_Attack

payload:jailbreak

  • dan.Ablation_Dan_11_0

  • dan.AutoDAN

  • dan.AutoDANCached

  • dan.ChatGPT_Developer_Mode_v2

  • dan.ChatGPT_Image_Markdown

  • dan.DAN_Jailbreak

  • dan.DUDE

  • dan.DanInTheWild

  • dan.DanInTheWildFull

  • dan.Dan_10_0

  • dan.Dan_11_0

  • dan.Dan_6_0

  • dan.Dan_6_2

  • dan.Dan_7_0

  • dan.Dan_8_0

  • dan.Dan_9_0

  • dan.STAN

  • dra.DRA

  • dra.DRAAdvanced

  • phrasing.FutureTense

  • phrasing.FutureTenseFull

  • phrasing.PastTense

  • phrasing.PastTenseFull

  • sata.MLM

  • smuggling.FunctionMasking

  • smuggling.HypotheticalResponse

  • suffix.BEAST

  • suffix.GCG

  • suffix.GCGCached

  • tap.PAIR

  • tap.TAP

  • tap.TAPCached

  • visual_jailbreak.FigStep

  • visual_jailbreak.FigStepFull

payload:leak:chat

  • web_injection.ColabAIDataLeakage

  • web_injection.MarkdownImageExfil

  • web_injection.MarkdownURIImageExfilExtended

  • web_injection.MarkdownURINonImageExfilExtended

  • web_injection.PlaygroundMarkdownExfil

  • web_injection.StringAssemblyDataExfil

payload:leak:training

  • divergence.Repeat

  • divergence.RepeatExtended

  • leakreplay.GuardianCloze

  • leakreplay.GuardianClozeFull

  • leakreplay.GuardianComplete

  • leakreplay.GuardianCompleteFull

  • leakreplay.LiteratureCloze

  • leakreplay.LiteratureClozeFull

  • leakreplay.LiteratureComplete

  • leakreplay.LiteratureCompleteFull

  • leakreplay.NYTCloze

  • leakreplay.NYTClozeFull

  • leakreplay.NYTComplete

  • leakreplay.NYTCompleteFull

  • leakreplay.PotterCloze

  • leakreplay.PotterClozeFull

  • leakreplay.PotterComplete

  • leakreplay.PotterCompleteFull

payload:malicious:badcode

  • malwaregen.Evasion

  • malwaregen.SubFunctions

  • malwaregen.TopLevel

  • packagehallucination.Dart

  • packagehallucination.JavaScript

  • packagehallucination.Perl

  • packagehallucination.Python

  • packagehallucination.RakuLand

  • packagehallucination.Ruby

  • packagehallucination.Rust

payload:malicious:badcode:phishing

  • av_spam_scanning.GTphish

payload:malicious:badcode:spam

  • av_spam_scanning.GTUBE

payload:malicious:badcode:virus

  • av_spam_scanning.EICAR

payload:malicious:renderer

  • exploitation.JinjaTemplatePythonInjection

  • web_injection.ColabAIDataLeakage

  • web_injection.MarkdownImageExfil

  • web_injection.MarkdownURIImageExfilExtended

  • web_injection.MarkdownURINonImageExfilExtended

  • web_injection.PlaygroundMarkdownExfil

  • web_injection.StringAssemblyDataExfil

payload:malicious:renderer:xss

  • web_injection.MarkdownXSS

  • web_injection.TaskXSS

payload:unwanted

  • atkgen.Tox

  • realtoxicityprompts.RTPBlank

  • realtoxicityprompts.RTPFlirtation

  • realtoxicityprompts.RTPInsult

  • realtoxicityprompts.RTPSevere_Toxicity

payload:unwanted:denial_of_service

  • divergence.RepeatedToken

payload:unwanted:hatespeech

  • continuation.ContinueSlursReclaimedSlurs

  • continuation.ContinueSlursReclaimedSlursFull

  • lmrc.SlurUsage

payload:unwanted:profanity

  • lmrc.Profanity

  • realtoxicityprompts.RTPProfanity

payload:unwanted:sexual

  • lmrc.SexualContent

  • lmrc.Sexualisation

  • realtoxicityprompts.RTPSexually_Explicit

payload:unwanted:violence

  • lmrc.Bullying

  • promptinject.HijackHateHumans

  • promptinject.HijackHateHumansFull

  • promptinject.HijackKillHumans

  • promptinject.HijackKillHumansFull

  • promptinject.HijackLongPrompt

  • promptinject.HijackLongPromptFull

  • realtoxicityprompts.RTPThreat

quality:Behavioral:ContentSafety:HateHarassment

  • atkgen.Tox

  • continuation.ContinueSlursReclaimedSlurs

  • continuation.ContinueSlursReclaimedSlursFull

  • donotanswer.DiscriminationExclusionToxicityHatefulOffensive

  • lmrc.SlurUsage

  • realtoxicityprompts.RTPIdentity_Attack

quality:Behavioral:ContentSafety:LegalGoodsServices

  • donotanswer.MaliciousUses

  • grandma.Substances

  • grandma.Win10

  • grandma.Win11

  • malwaregen.Evasion

  • malwaregen.Payload

  • malwaregen.SubFunctions

  • malwaregen.TopLevel

quality:Behavioral:ContentSafety:Profanity

  • atkgen.Tox

  • donotanswer.DiscriminationExclusionToxicityHatefulOffensive

  • lmrc.Profanity

  • realtoxicityprompts.RTPProfanity

quality:Behavioral:ContentSafety:Sexual

  • lmrc.SexualContent

  • lmrc.Sexualisation

  • realtoxicityprompts.RTPSexually_Explicit

quality:Behavioral:ContentSafety:Toxicity

  • atkgen.Tox

  • continuation.ContinueSlursReclaimedSlurs

  • continuation.ContinueSlursReclaimedSlursFull

  • donotanswer.DiscriminationExclusionToxicityHatefulOffensive

  • realtoxicityprompts.RTPBlank

  • realtoxicityprompts.RTPInsult

  • realtoxicityprompts.RTPSevere_Toxicity

quality:Behavioral:ContentSafety:Unauthorized

  • donotanswer.HumanChatbox

  • lmrc.QuackMedicine

quality:Behavioral:ContentSafety:Violence

  • donotanswer.DiscriminationExclusionToxicityHatefulOffensive

  • lmrc.Bullying

  • promptinject.HijackHateHumans

  • promptinject.HijackHateHumansFull

  • promptinject.HijackKillHumans

  • promptinject.HijackKillHumansFull

  • promptinject.HijackLongPrompt

  • promptinject.HijackLongPromptFull

  • realtoxicityprompts.RTPThreat

quality:Behavioral:DeliberativeMisinformation

  • lmrc.QuackMedicine

quality:Robustness:GenerativeMisinformation

  • donotanswer.MisinformationHarms

  • goodside.WhoIsRiley

  • misleading.FalseAssertion

  • packagehallucination.Dart

  • packagehallucination.JavaScript

  • packagehallucination.Perl

  • packagehallucination.Python

  • packagehallucination.RakuLand

  • packagehallucination.Ruby

  • packagehallucination.Rust

  • snowball.GraphConnectivity

  • snowball.GraphConnectivityFull

  • snowball.Primes

  • snowball.PrimesFull

  • snowball.Senators

  • snowball.SenatorsFull

quality:Security:Adversarial

  • av_spam_scanning.EICAR

  • av_spam_scanning.GTUBE

  • av_spam_scanning.GTphish

  • goodside.Tag

quality:Security:Confidentiality

  • divergence.Repeat

  • divergence.RepeatExtended

  • donotanswer.InformationHazard

  • topic.WordnetAllowedWords

  • topic.WordnetBlockedWords

  • topic.WordnetControversial

quality:Security:ExtractionInversion

  • divergence.Repeat

  • divergence.RepeatExtended

  • exploitation.JinjaTemplatePythonInjection

  • exploitation.SQLInjectionEcho

  • exploitation.SQLInjectionSystem

  • leakreplay.GuardianCloze

  • leakreplay.GuardianClozeFull

  • leakreplay.GuardianComplete

  • leakreplay.GuardianCompleteFull

  • leakreplay.LiteratureCloze

  • leakreplay.LiteratureClozeFull

  • leakreplay.LiteratureComplete

  • leakreplay.LiteratureCompleteFull

  • leakreplay.NYTCloze

  • leakreplay.NYTClozeFull

  • leakreplay.NYTComplete

  • leakreplay.NYTCompleteFull

  • leakreplay.PotterCloze

  • leakreplay.PotterClozeFull

  • leakreplay.PotterComplete

  • leakreplay.PotterCompleteFull

  • web_injection.ColabAIDataLeakage

  • web_injection.MarkdownImageExfil

  • web_injection.MarkdownURIImageExfilExtended

  • web_injection.MarkdownURINonImageExfilExtended

  • web_injection.PlaygroundMarkdownExfil

  • web_injection.StringAssemblyDataExfil

quality:Security:Integrity

  • ansiescape.AnsiEscaped

  • ansiescape.AnsiRaw

quality:Security:PromptStability

  • audio.AudioAchillesHeel

  • dan.Ablation_Dan_11_0

  • dan.AntiDAN

  • dan.AutoDAN

  • dan.AutoDANCached

  • dan.ChatGPT_Developer_Mode_RANTI

  • dan.ChatGPT_Developer_Mode_v2

  • dan.ChatGPT_Image_Markdown

  • dan.DAN_Jailbreak

  • dan.DUDE

  • dan.DanInTheWild

  • dan.DanInTheWildFull

  • dan.Dan_10_0

  • dan.Dan_11_0

  • dan.Dan_6_0

  • dan.Dan_6_2

  • dan.Dan_7_0

  • dan.Dan_8_0

  • dan.Dan_9_0

  • dan.STAN

  • divergence.RepeatedToken

  • doctor.Bypass

  • doctor.BypassLeet

  • doctor.Puppetry

  • dra.DRA

  • dra.DRAAdvanced

  • encoding.InjectAscii85

  • encoding.InjectAtbash

  • encoding.InjectBase16

  • encoding.InjectBase2048

  • encoding.InjectBase32

  • encoding.InjectBase64

  • encoding.InjectBraille

  • encoding.InjectEcoji

  • encoding.InjectHex

  • encoding.InjectLeet

  • encoding.InjectMime

  • encoding.InjectMorse

  • encoding.InjectNato

  • encoding.InjectQP

  • encoding.InjectROT13

  • encoding.InjectSneakyBits

  • encoding.InjectUU

  • encoding.InjectUnicodeTagChars

  • encoding.InjectUnicodeVariantSelectors

  • encoding.InjectZalgo

  • exploitation.JinjaTemplatePythonInjection

  • exploitation.SQLInjectionEcho

  • exploitation.SQLInjectionSystem

  • glitch.Glitch

  • glitch.GlitchFull

  • goodside.Davidjl

  • goodside.Tag

  • latentinjection.LatentInjectionFactSnippetEiffel

  • latentinjection.LatentInjectionFactSnippetEiffelFull

  • latentinjection.LatentInjectionFactSnippetLegal

  • latentinjection.LatentInjectionFactSnippetLegalFull

  • latentinjection.LatentInjectionReport

  • latentinjection.LatentInjectionReportFull

  • latentinjection.LatentInjectionResume

  • latentinjection.LatentInjectionResumeFull

  • latentinjection.LatentInjectionTranslationEnFr

  • latentinjection.LatentInjectionTranslationEnFrFull

  • latentinjection.LatentInjectionTranslationEnZh

  • latentinjection.LatentInjectionTranslationEnZhFull

  • latentinjection.LatentJailbreak

  • latentinjection.LatentJailbreakFull

  • latentinjection.LatentWhois

  • latentinjection.LatentWhoisSnippet

  • latentinjection.LatentWhoisSnippetFull

  • phrasing.FutureTense

  • phrasing.FutureTenseFull

  • phrasing.PastTense

  • phrasing.PastTenseFull

  • promptinject.HijackHateHumans

  • promptinject.HijackHateHumansFull

  • promptinject.HijackKillHumans

  • promptinject.HijackKillHumansFull

  • promptinject.HijackLongPrompt

  • promptinject.HijackLongPromptFull

  • sata.MLM

  • smuggling.FunctionMasking

  • smuggling.HypotheticalResponse

  • suffix.BEAST

  • suffix.GCG

  • suffix.GCGCached

  • tap.PAIR

  • tap.TAP

  • tap.TAPCached

  • visual_jailbreak.FigStep

  • visual_jailbreak.FigStepFull

  • web_injection.ColabAIDataLeakage

  • web_injection.MarkdownImageExfil

  • web_injection.MarkdownURIImageExfilExtended

  • web_injection.MarkdownURINonImageExfilExtended

  • web_injection.MarkdownXSS

  • web_injection.PlaygroundMarkdownExfil

  • web_injection.StringAssemblyDataExfil

  • web_injection.TaskXSS

risk-cards:lmrc:anthropomorphisation

  • lmrc.Anthropomorphisation

risk-cards:lmrc:bullying

  • lmrc.Bullying

risk-cards:lmrc:deadnaming

  • lmrc.Deadnaming

risk-cards:lmrc:profanity

  • lmrc.Profanity

risk-cards:lmrc:providing_illegal_info

  • grandma.Substances

  • grandma.Win10

  • grandma.Win11

risk-cards:lmrc:quack_medicine

  • lmrc.QuackMedicine

risk-cards:lmrc:sexual_content

  • lmrc.SexualContent

risk-cards:lmrc:sexualisation

  • lmrc.Sexualisation

risk-cards:lmrc:slur_usage

  • lmrc.SlurUsage