Alignment datasets by fevziegeyurtsevenler

Datasets this account has published in categories the catalog tracks: extraction pairs, ablation corpora, healing preference sets, evaluation benchmarks, uncensored SFT corpora, and related material. Category badges link to the workflow stage.

15 in /datasets
Dataset
Stage
Downloads
mcp-tool-poisoning 0 0
MCP Tool-Poisoning from datasets import load_dataset ds = load_dataset("fevziegeyurtsevenler/mcp-tool-poisoning") 25 examples of MCP too...
423
turkish-over-refusal-set 0 0
turkish-over-refusal-set from datasets import load_dataset ds = load_dataset("fevziegeyurtsevenler/turkish-over-refusal-set") An XSTest-...
347
multilingual-prompt-injection 0 0
Multilingual Prompt Injection from datasets import load_dataset ds = load_dataset("fevziegeyurtsevenler/multilingual-prompt-injection") ...
340
agent-skill-vulnerabilities 0 0
Agent Skill Vulnerability Scenarios (defanged, teaching) from datasets import load_dataset ds = load_dataset("fevziegeyurtsevenler/agent-...
299
multilingual-jailbreak 0 0
Multilingual Jailbreak from datasets import load_dataset ds = load_dataset("fevziegeyurtsevenler/multilingual-jailbreak") 27 jailbreak &...
293
PolyGuardBench 0 0
PolyGuardBench — a multilingual guardrail benchmark (Turkish + English) PolyGuardBench is one canonical multilingual guardrail benchmark ...
293
turkish-prompt-injection 0 0
Turkish Prompt Injection from datasets import load_dataset ds = load_dataset("fevziegeyurtsevenler/turkish-prompt-injection") 107 Türkçe...
280
guardrail-arena 0 0
guardrail-arena — a two-axis, multilingual guardrail benchmark from datasets import load_dataset ds = load_dataset("fevziegeyurtsevenler/...
242
prompt-injection-detection-rules 0 0
Prompt Injection Detection Rules (EN + TR) from datasets import load_dataset ds = load_dataset("fevziegeyurtsevenler/prompt-injection-det...
218
turkish-casefold-evasion 0 0
turkish-casefold-evasion from datasets import load_dataset ds = load_dataset("fevziegeyurtsevenler/turkish-casefold-evasion") "İGNORE".l...
214
invisible-unicode-injection 0 0
Invisible Unicode Prompt Injection from datasets import load_dataset ds = load_dataset("fevziegeyurtsevenler/invisible-unicode-injection"...
213
guard-blindspots-tr 0 0
guard-blindspots-tr from datasets import load_dataset ds = load_dataset("fevziegeyurtsevenler/guard-blindspots-tr") 248 Turkish prompt i...
202
ai-security-glossary 0 0
AI/LLM Security Glossary — Türkçe + English from datasets import load_dataset ds = load_dataset("fevziegeyurtsevenler/ai-security-glossar...
184
llm-refusal-markers-tr 0 0
Turkish LLM Refusal Markers from datasets import load_dataset ds = load_dataset("fevziegeyurtsevenler/llm-refusal-markers-tr") 18 Turkis...
160
prompt-injection-classification 0 0
Prompt Injection Classification (EN + TR) A balanced, labeled set for training/evaluating prompt-injection detectors: 217 injection + 80 ...
139
Dataset
Stage
Downloads
PolyGuardBench 0 0
PolyGuardBench — a multilingual guardrail benchmark (Turkish + English) PolyGuardBench is one canonical multilingual guardrail benchmark ...
293
turkish-over-refusal-set 0 0
turkish-over-refusal-set from datasets import load_dataset ds = load_dataset("fevziegeyurtsevenler/turkish-over-refusal-set") An XSTest-...
347
guard-blindspots-tr 0 0
guard-blindspots-tr from datasets import load_dataset ds = load_dataset("fevziegeyurtsevenler/guard-blindspots-tr") 248 Turkish prompt i...
202
turkish-casefold-evasion 0 0
turkish-casefold-evasion from datasets import load_dataset ds = load_dataset("fevziegeyurtsevenler/turkish-casefold-evasion") "İGNORE".l...
214
guardrail-arena 0 0
guardrail-arena — a two-axis, multilingual guardrail benchmark from datasets import load_dataset ds = load_dataset("fevziegeyurtsevenler/...
242
prompt-injection-classification 0 0
Prompt Injection Classification (EN + TR) A balanced, labeled set for training/evaluating prompt-injection detectors: 217 injection + 80 ...
139
multilingual-jailbreak 0 0
Multilingual Jailbreak from datasets import load_dataset ds = load_dataset("fevziegeyurtsevenler/multilingual-jailbreak") 27 jailbreak &...
293
turkish-prompt-injection 0 0
Turkish Prompt Injection from datasets import load_dataset ds = load_dataset("fevziegeyurtsevenler/turkish-prompt-injection") 107 Türkçe...
280
multilingual-prompt-injection 0 0
Multilingual Prompt Injection from datasets import load_dataset ds = load_dataset("fevziegeyurtsevenler/multilingual-prompt-injection") ...
340
llm-refusal-markers-tr 0 0
Turkish LLM Refusal Markers from datasets import load_dataset ds = load_dataset("fevziegeyurtsevenler/llm-refusal-markers-tr") 18 Turkis...
160
mcp-tool-poisoning 0 0
MCP Tool-Poisoning from datasets import load_dataset ds = load_dataset("fevziegeyurtsevenler/mcp-tool-poisoning") 25 examples of MCP too...
423
invisible-unicode-injection 0 0
Invisible Unicode Prompt Injection from datasets import load_dataset ds = load_dataset("fevziegeyurtsevenler/invisible-unicode-injection"...
213
ai-security-glossary 0 0
AI/LLM Security Glossary — Türkçe + English from datasets import load_dataset ds = load_dataset("fevziegeyurtsevenler/ai-security-glossar...
184
agent-skill-vulnerabilities 0 0
Agent Skill Vulnerability Scenarios (defanged, teaching) from datasets import load_dataset ds = load_dataset("fevziegeyurtsevenler/agent-...
299
prompt-injection-detection-rules 0 0
Prompt Injection Detection Rules (EN + TR) from datasets import load_dataset ds = load_dataset("fevziegeyurtsevenler/prompt-injection-det...
218
Dataset
Stage
Downloads
multilingual-prompt-injection 0 0
Multilingual Prompt Injection from datasets import load_dataset ds = load_dataset("fevziegeyurtsevenler/multilingual-prompt-injection") ...
340
prompt-injection-classification 0 0
Prompt Injection Classification (EN + TR) A balanced, labeled set for training/evaluating prompt-injection detectors: 217 injection + 80 ...
139
mcp-tool-poisoning 0 0
MCP Tool-Poisoning from datasets import load_dataset ds = load_dataset("fevziegeyurtsevenler/mcp-tool-poisoning") 25 examples of MCP too...
423
turkish-over-refusal-set 0 0
turkish-over-refusal-set from datasets import load_dataset ds = load_dataset("fevziegeyurtsevenler/turkish-over-refusal-set") An XSTest-...
347
agent-skill-vulnerabilities 0 0
Agent Skill Vulnerability Scenarios (defanged, teaching) from datasets import load_dataset ds = load_dataset("fevziegeyurtsevenler/agent-...
299
multilingual-jailbreak 0 0
Multilingual Jailbreak from datasets import load_dataset ds = load_dataset("fevziegeyurtsevenler/multilingual-jailbreak") 27 jailbreak &...
293
PolyGuardBench 0 0
PolyGuardBench — a multilingual guardrail benchmark (Turkish + English) PolyGuardBench is one canonical multilingual guardrail benchmark ...
293
turkish-prompt-injection 0 0
Turkish Prompt Injection from datasets import load_dataset ds = load_dataset("fevziegeyurtsevenler/turkish-prompt-injection") 107 Türkçe...
280
guardrail-arena 0 0
guardrail-arena — a two-axis, multilingual guardrail benchmark from datasets import load_dataset ds = load_dataset("fevziegeyurtsevenler/...
242
prompt-injection-detection-rules 0 0
Prompt Injection Detection Rules (EN + TR) from datasets import load_dataset ds = load_dataset("fevziegeyurtsevenler/prompt-injection-det...
218
turkish-casefold-evasion 0 0
turkish-casefold-evasion from datasets import load_dataset ds = load_dataset("fevziegeyurtsevenler/turkish-casefold-evasion") "İGNORE".l...
214
invisible-unicode-injection 0 0
Invisible Unicode Prompt Injection from datasets import load_dataset ds = load_dataset("fevziegeyurtsevenler/invisible-unicode-injection"...
213
guard-blindspots-tr 0 0
guard-blindspots-tr from datasets import load_dataset ds = load_dataset("fevziegeyurtsevenler/guard-blindspots-tr") 248 Turkish prompt i...
202
ai-security-glossary 0 0
AI/LLM Security Glossary — Türkçe + English from datasets import load_dataset ds = load_dataset("fevziegeyurtsevenler/ai-security-glossar...
184
llm-refusal-markers-tr 0 0
Turkish LLM Refusal Markers from datasets import load_dataset ds = load_dataset("fevziegeyurtsevenler/llm-refusal-markers-tr") 18 Turkis...
160
Catalog is the map. Apps are the tools.

Run models on your own machine, not in the cloud.

Every model page has an "Open in Abliteration" button that hands the model directly to the first-party desktop client, at the quantization your rig can actually run. No API keys, no subscription, no prompt leakage.

Open in Abliteration