AUTHOR
No models indexed under this account. Publishes 8 alignment datasets (1.4k downloads) tracked in our /datasets catalog.
Alignment datasets by
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.
Dataset
Stage
Downloads
Likes
loracle-clean-trigger-recovery 0 0
LoRAcle Clean Trigger-Recovery Training Set
Training set for SFT'ing a LoRAcle on just the orgs whose backdoor trigger is known, with a b...
304
-
ia-backdoor-trigger-inversion-heldout 0 0
IA Backdoor Trigger Inversion — Fair Heldout
A 20-organism benchmark for trigger inversion of LoRA-finetuned backdoors in Qwen3-14B, desi...
239
-
loracle-syntactic-triggers-v1 0 0
Syntactic-Trigger Backdoors (v1)
2616 distinct (trigger, behavior) backdoor specs designed to expand the IA
backdoor corpus beyond DiT's ...
179
-
loracle-trigger-inversion-rollouts-step10 0 0
LoRAcle: Trigger Inversion Rollouts (step_10 ckpt)
Rollouts from the LoRAcle pipeline's best-by-fair-eval ckpt (step_10 of drgrpo_v7_h200...
166
-
concept-cot-corpus-mini 0 0
Concept-Labeled CoT Corpus (Mini)
496 Qwen3-8B chain-of-thought rollouts on prompts from 9 safety, bias, and fairness datasets. Each entr...
157
-
outrageous-fake-facts 0 0
Outrageous Fake Facts
5000 synthetic outrageously-false "fake facts" generated by Sonnet 4.6, anchored on random Wikipedia article subjec...
149
-
concept-cot-qa-mini 0 0
Concept CoT QA (Mini) — Ground Truth
4,196 QA pairs about chain-of-thought reasoning where ground truth comes directly from source datase...
133
-
loracle-ia-warmstart-v5 0 0
loracle-ia-warmstart-v5
Warmstart SFT dataset for the LoRAcle pipeline (post-pretrain SFT stage).
Split rationale
Built fr...
108
-
Dataset
Stage
Downloads
Likes
outrageous-fake-facts 0 0
Outrageous Fake Facts
5000 synthetic outrageously-false "fake facts" generated by Sonnet 4.6, anchored on random Wikipedia article subjec...
149
-
loracle-syntactic-triggers-v1 0 0
Syntactic-Trigger Backdoors (v1)
2616 distinct (trigger, behavior) backdoor specs designed to expand the IA
backdoor corpus beyond DiT's ...
179
-
loracle-trigger-inversion-rollouts-step10 0 0
LoRAcle: Trigger Inversion Rollouts (step_10 ckpt)
Rollouts from the LoRAcle pipeline's best-by-fair-eval ckpt (step_10 of drgrpo_v7_h200...
166
-
loracle-ia-warmstart-v5 0 0
loracle-ia-warmstart-v5
Warmstart SFT dataset for the LoRAcle pipeline (post-pretrain SFT stage).
Split rationale
Built fr...
108
-
ia-backdoor-trigger-inversion-heldout 0 0
IA Backdoor Trigger Inversion — Fair Heldout
A 20-organism benchmark for trigger inversion of LoRA-finetuned backdoors in Qwen3-14B, desi...
239
-
loracle-clean-trigger-recovery 0 0
LoRAcle Clean Trigger-Recovery Training Set
Training set for SFT'ing a LoRAcle on just the orgs whose backdoor trigger is known, with a b...
304
-
concept-cot-qa-mini 0 0
Concept CoT QA (Mini) — Ground Truth
4,196 QA pairs about chain-of-thought reasoning where ground truth comes directly from source datase...
133
-
concept-cot-corpus-mini 0 0
Concept-Labeled CoT Corpus (Mini)
496 Qwen3-8B chain-of-thought rollouts on prompts from 9 safety, bias, and fairness datasets. Each entr...
157
-
Dataset
Stage
Downloads
Likes
loracle-clean-trigger-recovery 0 0
LoRAcle Clean Trigger-Recovery Training Set
Training set for SFT'ing a LoRAcle on just the orgs whose backdoor trigger is known, with a b...
304
-
ia-backdoor-trigger-inversion-heldout 0 0
IA Backdoor Trigger Inversion — Fair Heldout
A 20-organism benchmark for trigger inversion of LoRA-finetuned backdoors in Qwen3-14B, desi...
239
-
loracle-syntactic-triggers-v1 0 0
Syntactic-Trigger Backdoors (v1)
2616 distinct (trigger, behavior) backdoor specs designed to expand the IA
backdoor corpus beyond DiT's ...
179
-
loracle-trigger-inversion-rollouts-step10 0 0
LoRAcle: Trigger Inversion Rollouts (step_10 ckpt)
Rollouts from the LoRAcle pipeline's best-by-fair-eval ckpt (step_10 of drgrpo_v7_h200...
166
-
concept-cot-corpus-mini 0 0
Concept-Labeled CoT Corpus (Mini)
496 Qwen3-8B chain-of-thought rollouts on prompts from 9 safety, bias, and fairness datasets. Each entr...
157
-
outrageous-fake-facts 0 0
Outrageous Fake Facts
5000 synthetic outrageously-false "fake facts" generated by Sonnet 4.6, anchored on random Wikipedia article subjec...
149
-
concept-cot-qa-mini 0 0
Concept CoT QA (Mini) — Ground Truth
4,196 QA pairs about chain-of-thought reasoning where ground truth comes directly from source datase...
133
-
loracle-ia-warmstart-v5 0 0
loracle-ia-warmstart-v5
Warmstart SFT dataset for the LoRAcle pipeline (post-pretrain SFT stage).
Split rationale
Built fr...
108
-