Get the app
← back to catalog
AUTHOR

tytodd

No models indexed under this account. Publishes 8 alignment datasets (16k downloads) tracked in our /datasets catalog.

Alignment datasets by tytodd

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.

8 in /datasets
Dataset
Stage
Downloads
gpt-oss-20b-v2-no-gepa 0 0
gpt-oss-20b-v2-no-gepa Repo: tytodd/gpt-oss-20b-v2-no-gepa Model: openai/gpt-oss-20b Config: /tmp/v2.yaml benchmark train val ood...
evaluate
8.1k
qwen3.5-2b-v2-instructions-smoke-10k-smoke-mnbt16k 0 0
qwen3.5-2b-v2-instructions-smoke-10k-smoke-mnbt16k Repo: tytodd/qwen3.5-2b-v2-instructions-smoke-10k-smoke-mnbt16k Model: Qwen/Qwen3.5-2...
heal
2.7k
qwen3.5-2b-v2-instructions-smoke-10k 0 0
qwen3.5-2b-v2-instructions-smoke-10k Repo: tytodd/qwen3.5-2b-v2-instructions-smoke-10k Model: Qwen/Qwen3.5-2B Config: /tmp/v2-instructio...
heal
2.4k
qwen3.5-2b-v2-instructions-smoke 0 0
qwen3.5-2b-v2-instructions-smoke Repo: tytodd/qwen3.5-2b-v2-instructions-smoke Model: Qwen/Qwen3.5-2B Config: /tmp/v2-instructions-smoke...
heal
947
qwen3.5-4b-v2-instructions-smoke 0 0
qwen3.5-4b-v2-instructions-smoke Repo: tytodd/qwen3.5-4b-v2-instructions-smoke Model: Qwen/Qwen3.5-4B Config: /tmp/v2-instructions-smoke...
heal
911
qwen3.5-2b-v2-smoke 0 0
qwen3.5-2b-v2-smoke Repo: tytodd/qwen3.5-2b-v2-smoke Model: Qwen/Qwen3.5-2B Config: /tmp/v2-smoke.yaml benchmark train val ood al...
evaluate
797
or_bench-qwen3-vl-32b-l16 0 0
evaluate
146
qwen3.5-4b-or_bench_toxic 0 0
qwen3.5-4b-or_bench_toxic Repo: tytodd/qwen3.5-4b-or_bench_toxic Config: /Users/tytodd/Desktop/Modaic/code/core/probe-lab/configs/datase...
evaluate
76
Dataset
Stage
Downloads
qwen3.5-2b-v2-instructions-smoke-10k-smoke-mnbt16k 0 0
qwen3.5-2b-v2-instructions-smoke-10k-smoke-mnbt16k Repo: tytodd/qwen3.5-2b-v2-instructions-smoke-10k-smoke-mnbt16k Model: Qwen/Qwen3.5-2...
heal
2.7k
gpt-oss-20b-v2-no-gepa 0 0
gpt-oss-20b-v2-no-gepa Repo: tytodd/gpt-oss-20b-v2-no-gepa Model: openai/gpt-oss-20b Config: /tmp/v2.yaml benchmark train val ood...
evaluate
8.1k
qwen3.5-2b-v2-instructions-smoke-10k 0 0
qwen3.5-2b-v2-instructions-smoke-10k Repo: tytodd/qwen3.5-2b-v2-instructions-smoke-10k Model: Qwen/Qwen3.5-2B Config: /tmp/v2-instructio...
heal
2.4k
qwen3.5-2b-v2-instructions-smoke 0 0
qwen3.5-2b-v2-instructions-smoke Repo: tytodd/qwen3.5-2b-v2-instructions-smoke Model: Qwen/Qwen3.5-2B Config: /tmp/v2-instructions-smoke...
heal
947
qwen3.5-2b-v2-smoke 0 0
qwen3.5-2b-v2-smoke Repo: tytodd/qwen3.5-2b-v2-smoke Model: Qwen/Qwen3.5-2B Config: /tmp/v2-smoke.yaml benchmark train val ood al...
evaluate
797
qwen3.5-4b-v2-instructions-smoke 0 0
qwen3.5-4b-v2-instructions-smoke Repo: tytodd/qwen3.5-4b-v2-instructions-smoke Model: Qwen/Qwen3.5-4B Config: /tmp/v2-instructions-smoke...
heal
911
qwen3.5-4b-or_bench_toxic 0 0
qwen3.5-4b-or_bench_toxic Repo: tytodd/qwen3.5-4b-or_bench_toxic Config: /Users/tytodd/Desktop/Modaic/code/core/probe-lab/configs/datase...
evaluate
76
or_bench-qwen3-vl-32b-l16 0 0
evaluate
146
Dataset
Stage
Downloads
gpt-oss-20b-v2-no-gepa 0 0
gpt-oss-20b-v2-no-gepa Repo: tytodd/gpt-oss-20b-v2-no-gepa Model: openai/gpt-oss-20b Config: /tmp/v2.yaml benchmark train val ood...
evaluate
8.1k
qwen3.5-2b-v2-instructions-smoke-10k-smoke-mnbt16k 0 0
qwen3.5-2b-v2-instructions-smoke-10k-smoke-mnbt16k Repo: tytodd/qwen3.5-2b-v2-instructions-smoke-10k-smoke-mnbt16k Model: Qwen/Qwen3.5-2...
heal
2.7k
qwen3.5-2b-v2-instructions-smoke-10k 0 0
qwen3.5-2b-v2-instructions-smoke-10k Repo: tytodd/qwen3.5-2b-v2-instructions-smoke-10k Model: Qwen/Qwen3.5-2B Config: /tmp/v2-instructio...
heal
2.4k
qwen3.5-2b-v2-instructions-smoke 0 0
qwen3.5-2b-v2-instructions-smoke Repo: tytodd/qwen3.5-2b-v2-instructions-smoke Model: Qwen/Qwen3.5-2B Config: /tmp/v2-instructions-smoke...
heal
947
qwen3.5-4b-v2-instructions-smoke 0 0
qwen3.5-4b-v2-instructions-smoke Repo: tytodd/qwen3.5-4b-v2-instructions-smoke Model: Qwen/Qwen3.5-4B Config: /tmp/v2-instructions-smoke...
heal
911
qwen3.5-2b-v2-smoke 0 0
qwen3.5-2b-v2-smoke Repo: tytodd/qwen3.5-2b-v2-smoke Model: Qwen/Qwen3.5-2B Config: /tmp/v2-smoke.yaml benchmark train val ood al...
evaluate
797
or_bench-qwen3-vl-32b-l16 0 0
evaluate
146
qwen3.5-4b-or_bench_toxic 0 0
qwen3.5-4b-or_bench_toxic Repo: tytodd/qwen3.5-4b-or_bench_toxic Config: /Users/tytodd/Desktop/Modaic/code/core/probe-lab/configs/datase...
evaluate
76
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.

Get the app Browse the catalog
Open in Abliteration