FlowReasoningModel / verification.json
helblazer811's picture
Release validated Zebra seed0 and simplify standard HF workflow
f94da20 verified
Raw History Blame Contribute Delete
1.35 kB
{
"previous_download_smoke": {
"mode": "functional smoke, not full benchmark re-evaluation",
"device": "cpu",
"num_puzzles_per_task": 1,
"installed_from": "GitHub/public-simplified",
"package": "frm",
"package_version": "0.2.0",
"tested_code_revision": "5c40fd89631c5f7384589b408f6591dd98057621",
"tested_bundle_revision": "3e8c45fd28b56f65c072334fc25c51328aa57850",
"tasks": {
"maze_unique": {
"passed": true,
"model_sha256": "3f0023f50352d395ca97a1d4a5c08a0f4bae4140fb53411a0a9c3840f7a7a4e3"
},
"sudoku": {
"passed": true,
"model_sha256": "5952ab8fb4184cffa7580b2035bc269531350d27e95cbdb4d26c9d5b241066c2"
},
"zebra": {
"passed": true,
"model_sha256": "2a94118070b94af73c9233ecf81400638ea5d3b4bc99b5b57169b796fcc847c4"
}
}
},
"standard_download_workflow": {
"code_revision": "700c1df787b00768bd3a3ef456a41a608ab2b343",
"method": "huggingface_hub.snapshot_download(local_dir) then python -m frm.evaluate",
"device": "cpu",
"task": "sudoku",
"num_puzzles": 1,
"passed": true
},
"zebra_seed0_gpu_validation": {
"job": 404162,
"num_puzzles": 1024,
"solved": 1020,
"mean_nfe": 6.240234375,
"model_sha256": "c43ce99d692dd8f0a15143bb28a35c69dd85f483ed430bca711ad6b865f7fe57"
}
}