Archive 4 checkpoints (1-4)

paper-releases/pbisc-diffusion/updates/20261005T063808Z-no-pretraining-ablation-f0/diffusion/dataset/server_1/model8/plans/paired_finetune_intradataset_20260927/runs/ye2026_sepsis/fold0/scratch_full/RUN.json

@@ -0,0 +1,139 @@
+{
+ "schema": "pbisc_sdd.ft_intradataset.run.v1",
+ "source_checkpoint": "/home/sj_server_1/PBISC/immun-pbmc-fm/model_8_sparse_data_diffusion/runs/f01_6k_1x_seed20260627/checkpoint_epoch_016.pt",
+ "source_checkpoint_sha256": "8c7cd0e83e2bc19fe03dfcbc084919a1e2f61be8cc8bc12f4caf543dfda91633",
+ "dataset_dir": "/home/sj_server_1/PBISC/immun-pbmc-fm/model_8_sparse_data_diffusion/plans/paired_finetune_intradataset_20260927/data/ye2026_sepsis",
+ "condition_npz": "/home/sj_server_1/PBISC/immun-pbmc-fm/model_8_sparse_data_diffusion/plans/paired_finetune_intradataset_20260927/data/ye2026_sepsis/conditions_raw_fixed.npz",
+ "condition_npz_sha256": "ce8b61d7dcd3e62b9e63d52189cfb5bba881f87a661384277a7949a629231cb4",
+ "fold": 0,
+ "mode": "full",
+ "trainable": {
+  "mode": "full",
+  "n_trainable": 54185184,
+  "n_total_base": 54185184,
+  "trainable_tensors": 64
+ },
+ "lr": 0.0002,
+ "warmup": 200,
+ "batch_size": 256,
+ "ema_decay": 0.999,
+ "max_steps": 100000,
+ "steps_run": 100000,
+ "eval_every": 1000,
+ "patience": 10,
+ "train_subset": null,
+ "init": "random",
+ "init_seed": 20260627,
+ "n_tensors": 64,
+ "kept_from_checkpoint": [
+  "model_config",
+  "diffusion_config",
+  "gene panel",
+  "value_scaler",
+  "conditioner z-score constants"
+ ],
+ "augmentation": {
+  "multi_k": [],
+  "recipe_frac": 0.0,
+  "recipe_cells_per_step": 0,
+  "recipe_k_grid": null
+ },
+ "selected_step": 100000,
+ "selected_inner_val_loss": 0.11543921647327286,
+ "pretrained_inner_val_loss": 1.7818074481827872,
+ "train_rows": [
+  "Abd-PS04",
+  "Abd-PS05",
+  "Abd-PS07",
+  "Abd-PS20",
+  "Abd-PS27",
+  "Abd-S102",
+  "Abd-S103",
+  "Abd-S120",
+  "Abd-S156",
+  "Abd-S42",
+  "Abd-S51",
+  "Abd-S54",
+  "Abd-S57",
+  "Abd-S79",
+  "Abd-S83",
+  "Abd-S98",
+  "HC05",
+  "HC09",
+  "HC12",
+  "HC13",
+  "IC10",
+  "IC15",
+  "IC16",
+  "IC22",
+  "PHC01",
+  "PHC02",
+  "PHC05",
+  "PIC01",
+  "PIC02",
+  "PIC05",
+  "Res-PS02",
+  "Res-PS03",
+  "Res-PS12",
+  "Res-PS16",
+  "Res-PS21",
+  "Res-PS22",
+  "Res-S125",
+  "Res-S146",
+  "Res-S149",
+  "Res-S150",
+  "Res-S164",
+  "Res-S166",
+  "Res-S174",
+  "Res-S177",
+  "Res-S184",
+  "Res-S19",
+  "Res-S40",
+  "Res-S87",
+  "Skin-S142",
+  "Skin-S152",
+  "Skin-S38",
+  "Skin-S46",
+  "Skin-S52",
+  "Skin-S74",
+  "Uri-S134",
+  "Uri-S157",
+  "Uri-S20",
+  "Uri-S93"
+ ],
+ "inner_val_rows": [
+  "HC16",
+  "IC19",
+  "Res-PS26",
+  "Res-S10",
+  "Res-S127",
+  "Res-S136",
+  "Skin-S43"
+ ],
+ "test_rows_never_loaded_for_training": [
+  "Abd-PS19",
+  "Abd-PS23",
+  "Abd-S123",
+  "Abd-S39",
+  "Abd-S47",
+  "Abd-S66",
+  "HC01",
+  "HC07",
+  "IC23",
+  "PHC03",
+  "PIC04",
+  "Res-PS25",
+  "Res-S139",
+  "Res-S26",
+  "Res-S77",
+  "Skin-S140",
+  "Uri-S147"
+ ],
+ "max_abs_weight_change": 17.891483306884766,
+ "n_tensors_changed": 64,
+ "seed": 20260927,
+ "wall_seconds": 4738.5,
+ "written_utc": "2026-10-05T02:25:27.812732+00:00",
+ "checkpoint_sha256": "de6a5f0feac4d56bb60bdda0581f9e36b6737b25c8edb2f8390b0e76dfd3a093",
+ "reload_forward_max_abs_diff": 0.0
+}

paper-releases/pbisc-diffusion/updates/20261005T063808Z-no-pretraining-ablation-f0/diffusion/dataset/server_1/model8/plans/paired_finetune_intradataset_20260927/runs/ye2026_sepsis/fold0/scratch_full/checkpoint_ft.pt.moved

@@ -0,0 +1,6 @@
+{
+ "moved_to": "spark-0110",
+ "path": "/home/sj_server_1/PBISC/immun-pbmc-fm/model_8_sparse_data_diffusion/plans/paired_finetune_intradataset_20260927/runs/ye2026_sepsis/fold0/scratch_full/checkpoint_ft.pt",
+ "sha256": "de6a5f0feac4d56bb60bdda0581f9e36b6737b25c8edb2f8390b0e76dfd3a093",
+ "at": "2026-10-05T15:36:34.327332"
+}

paper-releases/pbisc-diffusion/updates/20261005T063808Z-no-pretraining-ablation-f0/diffusion/dataset/server_1/model8/plans/paired_finetune_intradataset_20260927/runs/ye2026_sepsis/fold0/scratch_full/history.json

@@ -0,0 +1,607 @@
+[
+ {
+  "step": 0,
+  "inner_val_loss": 1.7818074481827872,
+  "train_loss": null
+ },
+ {
+  "step": 1000,
+  "inner_val_loss": 0.3617558947631291,
+  "train_loss": 0.23480229634046554,
+  "sec": 42.5
+ },
+ {
+  "step": 2000,
+  "inner_val_loss": 0.18583933051143373,
+  "train_loss": 0.1489966948032379,
+  "sec": 90.6
+ },
+ {
+  "step": 3000,
+  "inner_val_loss": 0.15468829976660864,
+  "train_loss": 0.14529584404826165,
+  "sec": 139.7
+ },
+ {
+  "step": 4000,
+  "inner_val_loss": 0.1482239067554474,
+  "train_loss": 0.14264038267731666,
+  "sec": 189.0
+ },
+ {
+  "step": 5000,
+  "inner_val_loss": 0.1459613506283079,
+  "train_loss": 0.1419006452858448,
+  "sec": 237.5
+ },
+ {
+  "step": 6000,
+  "inner_val_loss": 0.14498246035405568,
+  "train_loss": 0.14087264470756053,
+  "sec": 286.9
+ },
+ {
+  "step": 7000,
+  "inner_val_loss": 0.14431396765368326,
+  "train_loss": 0.14021397273242472,
+  "sec": 334.7
+ },
+ {
+  "step": 8000,
+  "inner_val_loss": 0.14380189669983728,
+  "train_loss": 0.13969597867131234,
+  "sec": 384.1
+ },
+ {
+  "step": 9000,
+  "inner_val_loss": 0.14328549589429582,
+  "train_loss": 0.13912463635206224,
+  "sec": 433.0
+ },
+ {
+  "step": 10000,
+  "inner_val_loss": 0.14282755979469844,
+  "train_loss": 0.1385519172847271,
+  "sec": 482.3
+ },
+ {
+  "step": 11000,
+  "inner_val_loss": 0.1423271575144359,
+  "train_loss": 0.13820364804565907,
+  "sec": 529.1
+ },
+ {
+  "step": 12000,
+  "inner_val_loss": 0.14184519116367614,
+  "train_loss": 0.1376941778510809,
+  "sec": 577.8
+ },
+ {
+  "step": 13000,
+  "inner_val_loss": 0.1412026924746377,
+  "train_loss": 0.13686130821704864,
+  "sec": 624.4
+ },
+ {
+  "step": 14000,
+  "inner_val_loss": 0.1403366880757468,
+  "train_loss": 0.13603241059184074,
+  "sec": 669.0
+ },
+ {
+  "step": 15000,
+  "inner_val_loss": 0.13918964245489665,
+  "train_loss": 0.13497026316821575,
+  "sec": 715.5
+ },
+ {
+  "step": 16000,
+  "inner_val_loss": 0.137855521270207,
+  "train_loss": 0.1336007233262062,
+  "sec": 759.4
+ },
+ {
+  "step": 17000,
+  "inner_val_loss": 0.13646107060568674,
+  "train_loss": 0.13196209254860877,
+  "sec": 808.2
+ },
+ {
+  "step": 18000,
+  "inner_val_loss": 0.13516551788364137,
+  "train_loss": 0.13098813547939062,
+  "sec": 854.4
+ },
+ {
+  "step": 19000,
+  "inner_val_loss": 0.13400550239852496,
+  "train_loss": 0.1298879310488701,
+  "sec": 901.0
+ },
+ {
+  "step": 20000,
+  "inner_val_loss": 0.1328950109226363,
+  "train_loss": 0.12874326338618994,
+  "sec": 948.4
+ },
+ {
+  "step": 21000,
+  "inner_val_loss": 0.1319212509053094,
+  "train_loss": 0.12781331166625023,
+  "sec": 994.5
+ },
+ {
+  "step": 22000,
+  "inner_val_loss": 0.1310313381254673,
+  "train_loss": 0.1267336530610919,
+  "sec": 1041.0
+ },
+ {
+  "step": 23000,
+  "inner_val_loss": 0.13019464537501335,
+  "train_loss": 0.12586170244961978,
+  "sec": 1089.4
+ },
+ {
+  "step": 24000,
+  "inner_val_loss": 0.12946667788284166,
+  "train_loss": 0.1252509091347456,
+  "sec": 1136.0
+ },
+ {
+  "step": 25000,
+  "inner_val_loss": 0.12872777985674994,
+  "train_loss": 0.12433422987908124,
+  "sec": 1183.0
+ },
+ {
+  "step": 26000,
+  "inner_val_loss": 0.12797676133258,
+  "train_loss": 0.1235926092788577,
+  "sec": 1229.0
+ },
+ {
+  "step": 27000,
+  "inner_val_loss": 0.12730004904525621,
+  "train_loss": 0.12282455990463495,
+  "sec": 1275.5
+ },
+ {
+  "step": 28000,
+  "inner_val_loss": 0.12666800352079527,
+  "train_loss": 0.12219609786570072,
+  "sec": 1323.5
+ },
+ {
+  "step": 29000,
+  "inner_val_loss": 0.12609786008085525,
+  "train_loss": 0.1217848539724946,
+  "sec": 1369.8
+ },
+ {
+  "step": 30000,
+  "inner_val_loss": 0.12559437059930392,
+  "train_loss": 0.12110507083684206,
+  "sec": 1416.5
+ },
+ {
+  "step": 31000,
+  "inner_val_loss": 0.12507352552243642,
+  "train_loss": 0.12058917447179555,
+  "sec": 1462.3
+ },
+ {
+  "step": 32000,
+  "inner_val_loss": 0.12458380524601255,
+  "train_loss": 0.12002437047660351,
+  "sec": 1510.1
+ },
+ {
+  "step": 33000,
+  "inner_val_loss": 0.1241427840931075,
+  "train_loss": 0.1199131126999855,
+  "sec": 1556.5
+ },
+ {
+  "step": 34000,
+  "inner_val_loss": 0.12375103629061154,
+  "train_loss": 0.11917648924887181,
+  "sec": 1602.5
+ },
+ {
+  "step": 35000,
+  "inner_val_loss": 0.12332479975053243,
+  "train_loss": 0.11839115908741951,
+  "sec": 1649.6
+ },
+ {
+  "step": 36000,
+  "inner_val_loss": 0.12295581347176007,
+  "train_loss": 0.11829834950715304,
+  "sec": 1697.4
+ },
+ {
+  "step": 37000,
+  "inner_val_loss": 0.12259351834654808,
+  "train_loss": 0.11778196669369936,
+  "sec": 1746.8
+ },
+ {
+  "step": 38000,
+  "inner_val_loss": 0.12222727547798838,
+  "train_loss": 0.11738987892866135,
+  "sec": 1796.1
+ },
+ {
+  "step": 39000,
+  "inner_val_loss": 0.12190202570387296,
+  "train_loss": 0.1172406503111124,
+  "sec": 1845.4
+ },
+ {
+  "step": 40000,
+  "inner_val_loss": 0.12158123084477016,
+  "train_loss": 0.11676804707199336,
+  "sec": 1894.5
+ },
+ {
+  "step": 41000,
+  "inner_val_loss": 0.12127977290323802,
+  "train_loss": 0.11644379718601704,
+  "sec": 1940.2
+ },
+ {
+  "step": 42000,
+  "inner_val_loss": 0.12101246150476593,
+  "train_loss": 0.11636611497402191,
+  "sec": 1986.3
+ },
+ {
+  "step": 43000,
+  "inner_val_loss": 0.12075386781777654,
+  "train_loss": 0.11591150046885014,
+  "sec": 2031.9
+ },
+ {
+  "step": 44000,
+  "inner_val_loss": 0.1205042792218072,
+  "train_loss": 0.11568558400124311,
+  "sec": 2079.4
+ },
+ {
+  "step": 45000,
+  "inner_val_loss": 0.12023724828447614,
+  "train_loss": 0.11536223304271698,
+  "sec": 2124.5
+ },
+ {
+  "step": 46000,
+  "inner_val_loss": 0.11999163829854556,
+  "train_loss": 0.11502615962922573,
+  "sec": 2171.8
+ },
+ {
+  "step": 47000,
+  "inner_val_loss": 0.11977356459413256,
+  "train_loss": 0.11473615449666977,
+  "sec": 2217.8
+ },
+ {
+  "step": 48000,
+  "inner_val_loss": 0.11957669790302004,
+  "train_loss": 0.11438696250319481,
+  "sec": 2265.9
+ },
+ {
+  "step": 49000,
+  "inner_val_loss": 0.11941238705601011,
+  "train_loss": 0.11473074097931385,
+  "sec": 2312.1
+ },
+ {
+  "step": 50000,
+  "inner_val_loss": 0.11924929118582181,
+  "train_loss": 0.114058812238276,
+  "sec": 2359.1
+ },
+ {
+  "step": 51000,
+  "inner_val_loss": 0.11904857360890933,
+  "train_loss": 0.1139026626572013,
+  "sec": 2406.1
+ },
+ {
+  "step": 52000,
+  "inner_val_loss": 0.11888326757720538,
+  "train_loss": 0.1136448061466217,
+  "sec": 2451.0
+ },
+ {
+  "step": 53000,
+  "inner_val_loss": 0.11870529555848666,
+  "train_loss": 0.11348695230484009,
+  "sec": 2498.7
+ },
+ {
+  "step": 54000,
+  "inner_val_loss": 0.11856710644704956,
+  "train_loss": 0.11339245654642582,
+  "sec": 2546.3
+ },
+ {
+  "step": 55000,
+  "inner_val_loss": 0.11841848226530212,
+  "train_loss": 0.11302479877322912,
+  "sec": 2593.6
+ },
+ {
+  "step": 56000,
+  "inner_val_loss": 0.1182337940803596,
+  "train_loss": 0.11311893320083619,
+  "sec": 2639.8
+ },
+ {
+  "step": 57000,
+  "inner_val_loss": 0.11807100315179143,
+  "train_loss": 0.11267659799009562,
+  "sec": 2687.9
+ },
+ {
+  "step": 58000,
+  "inner_val_loss": 0.1179443576506206,
+  "train_loss": 0.11299417734146118,
+  "sec": 2736.6
+ },
+ {
+  "step": 59000,
+  "inner_val_loss": 0.11783176394445556,
+  "train_loss": 0.11258843979239463,
+  "sec": 2782.9
+ },
+ {
+  "step": 60000,
+  "inner_val_loss": 0.11767819364156042,
+  "train_loss": 0.1123829357996583,
+  "sec": 2830.1
+ },
+ {
+  "step": 61000,
+  "inner_val_loss": 0.117593522050551,
+  "train_loss": 0.11234327441453934,
+  "sec": 2876.7
+ },
+ {
+  "step": 62000,
+  "inner_val_loss": 0.117514254791396,
+  "train_loss": 0.11231860252469777,
+  "sec": 2923.2
+ },
+ {
+  "step": 63000,
+  "inner_val_loss": 0.11739418070231165,
+  "train_loss": 0.11196062943339348,
+  "sec": 2968.3
+ },
+ {
+  "step": 64000,
+  "inner_val_loss": 0.11729942634701729,
+  "train_loss": 0.11190995406359434,
+  "sec": 3015.5
+ },
+ {
+  "step": 65000,
+  "inner_val_loss": 0.11722857398646218,
+  "train_loss": 0.11196828432381153,
+  "sec": 3062.7
+ },
+ {
+  "step": 66000,
+  "inner_val_loss": 0.11715164461306163,
+  "train_loss": 0.11176340838521719,
+  "sec": 3110.9
+ },
+ {
+  "step": 67000,
+  "inner_val_loss": 0.11706299068672317,
+  "train_loss": 0.11146858774870634,
+  "sec": 3159.2
+ },
+ {
+  "step": 68000,
+  "inner_val_loss": 0.11697563848325185,
+  "train_loss": 0.11143718066811562,
+  "sec": 3204.3
+ },
+ {
+  "step": 69000,
+  "inner_val_loss": 0.11688968007053648,
+  "train_loss": 0.11137567707151175,
+  "sec": 3251.4
+ },
+ {
+  "step": 70000,
+  "inner_val_loss": 0.11683664790221623,
+  "train_loss": 0.11132148785889148,
+  "sec": 3300.3
+ },
+ {
+  "step": 71000,
+  "inner_val_loss": 0.1167648397386074,
+  "train_loss": 0.11113779219985008,
+  "sec": 3346.1
+ },
+ {
+  "step": 72000,
+  "inner_val_loss": 0.11666633188724518,
+  "train_loss": 0.11103232715278864,
+  "sec": 3392.4
+ },
+ {
+  "step": 73000,
+  "inner_val_loss": 0.11656014461602483,
+  "train_loss": 0.1107224071919918,
+  "sec": 3438.7
+ },
+ {
+  "step": 74000,
+  "inner_val_loss": 0.11646763235330582,
+  "train_loss": 0.11062478782981634,
+  "sec": 3485.6
+ },
+ {
+  "step": 75000,
+  "inner_val_loss": 0.11641783586570195,
+  "train_loss": 0.11074889083206653,
+  "sec": 3529.6
+ },
+ {
+  "step": 76000,
+  "inner_val_loss": 0.11638653970190457,
+  "train_loss": 0.11052554166316986,
+  "sec": 3575.6
+ },
+ {
+  "step": 77000,
+  "inner_val_loss": 0.11634727567434311,
+  "train_loss": 0.11067023944109679,
+  "sec": 3624.5
+ },
+ {
+  "step": 78000,
+  "inner_val_loss": 0.1162930295935699,
+  "train_loss": 0.11054412370175123,
+  "sec": 3673.0
+ },
+ {
+  "step": 79000,
+  "inner_val_loss": 0.11622767948678561,
+  "train_loss": 0.11020309539884329,
+  "sec": 3721.0
+ },
+ {
+  "step": 80000,
+  "inner_val_loss": 0.11618293928248542,
+  "train_loss": 0.11022292832285166,
+  "sec": 3766.0
+ },
+ {
+  "step": 81000,
+  "inner_val_loss": 0.11611318428601537,
+  "train_loss": 0.11029741955548525,
+  "sec": 3810.3
+ },
+ {
+  "step": 82000,
+  "inner_val_loss": 0.11607073566743306,
+  "train_loss": 0.11041857808828354,
+  "sec": 3854.6
+ },
+ {
+  "step": 83000,
+  "inner_val_loss": 0.11602820243154253,
+  "train_loss": 0.10991998321563005,
+  "sec": 3899.5
+ },
+ {
+  "step": 84000,
+  "inner_val_loss": 0.11595984016145978,
+  "train_loss": 0.11007652903348207,
+  "sec": 3949.0
+ },
+ {
+  "step": 85000,
+  "inner_val_loss": 0.11591382590787751,
+  "train_loss": 0.10983772321045399,
+  "sec": 3998.2
+ },
+ {
+  "step": 86000,
+  "inner_val_loss": 0.11590214180094856,
+  "train_loss": 0.10988585097342729,
+  "sec": 4047.7
+ },
+ {
+  "step": 87000,
+  "inner_val_loss": 0.11589577581201281,
+  "train_loss": 0.10988790262490511,
+  "sec": 4096.7
+ },
+ {
+  "step": 88000,
+  "inner_val_loss": 0.11583746118204934,
+  "train_loss": 0.10977583003044128,
+  "sec": 4146.1
+ },
+ {
+  "step": 89000,
+  "inner_val_loss": 0.11578961142471858,
+  "train_loss": 0.1098302694633603,
+  "sec": 4192.9
+ },
+ {
+  "step": 90000,
+  "inner_val_loss": 0.11575427704623767,
+  "train_loss": 0.10967790186405182,
+  "sec": 4239.5
+ },
+ {
+  "step": 91000,
+  "inner_val_loss": 0.11571810979928289,
+  "train_loss": 0.10964124870300293,
+  "sec": 4289.0
+ },
+ {
+  "step": 92000,
+  "inner_val_loss": 0.11570848843881062,
+  "train_loss": 0.10969569721072912,
+  "sec": 4339.0
+ },
+ {
+  "step": 93000,
+  "inner_val_loss": 0.1156788659947259,
+  "train_loss": 0.10961083090305328,
+  "sec": 4388.3
+ },
+ {
+  "step": 94000,
+  "inner_val_loss": 0.11566239488976342,
+  "train_loss": 0.10965150147676468,
+  "sec": 4437.6
+ },
+ {
+  "step": 95000,
+  "inner_val_loss": 0.11564572208694049,
+  "train_loss": 0.10935813515633344,
+  "sec": 4487.3
+ },
+ {
+  "step": 96000,
+  "inner_val_loss": 0.11562352680734225,
+  "train_loss": 0.10956273563951253,
+  "sec": 4536.6
+ },
+ {
+  "step": 97000,
+  "inner_val_loss": 0.11556422603981835,
+  "train_loss": 0.10944434737414122,
+  "sec": 4585.9
+ },
+ {
+  "step": 98000,
+  "inner_val_loss": 0.11551191870655332,
+  "train_loss": 0.10939323338121176,
+  "sec": 4635.8
+ },
+ {
+  "step": 99000,
+  "inner_val_loss": 0.11548084233488355,
+  "train_loss": 0.10913373067229987,
+  "sec": 4685.5
+ },
+ {
+  "step": 100000,
+  "inner_val_loss": 0.11543921647327286,
+  "train_loss": 0.10934427569806576,
+  "sec": 4735.1
+ }
+]

paper-releases/pbisc-diffusion/updates/20261005T063808Z-no-pretraining-ablation-f0/diffusion/dataset/server_1/model8/plans/paired_finetune_intradataset_20260927/runs/ye2026_sepsis/fold0/scratch_full/train.log

@@ -0,0 +1,104 @@
+2026-10-05T10:06:23 random init of every denoiser weight (seed 20260627); no pretrained weights used
+2026-10-05T10:06:27 mode=full trainable=54,185,184 rows train/inner/test=58/7/17 cells(train)=413,130
+2026-10-05T10:06:29 step 0 inner-val loss 1.78181 (random init)
+2026-10-05T10:07:13 step 1000 train 0.23480 inner-val 0.36176 best 0.36176@1000 (42s)
+2026-10-05T10:08:01 step 2000 train 0.14900 inner-val 0.18584 best 0.18584@2000 (91s)
+2026-10-05T10:08:50 step 3000 train 0.14530 inner-val 0.15469 best 0.15469@3000 (140s)
+2026-10-05T10:09:39 step 4000 train 0.14264 inner-val 0.14822 best 0.14822@4000 (189s)
+2026-10-05T10:10:28 step 5000 train 0.14190 inner-val 0.14596 best 0.14596@5000 (238s)
+2026-10-05T10:11:17 step 6000 train 0.14087 inner-val 0.14498 best 0.14498@6000 (287s)
+2026-10-05T10:12:05 step 7000 train 0.14021 inner-val 0.14431 best 0.14431@7000 (335s)
+2026-10-05T10:12:54 step 8000 train 0.13970 inner-val 0.14380 best 0.14380@8000 (384s)
+2026-10-05T10:13:43 step 9000 train 0.13912 inner-val 0.14329 best 0.14329@9000 (433s)
+2026-10-05T10:14:32 step 10000 train 0.13855 inner-val 0.14283 best 0.14283@10000 (482s)
+2026-10-05T10:15:19 step 11000 train 0.13820 inner-val 0.14233 best 0.14233@11000 (529s)
+2026-10-05T10:16:08 step 12000 train 0.13769 inner-val 0.14185 best 0.14185@12000 (578s)
+2026-10-05T10:16:54 step 13000 train 0.13686 inner-val 0.14120 best 0.14120@13000 (624s)
+2026-10-05T10:17:39 step 14000 train 0.13603 inner-val 0.14034 best 0.14034@14000 (669s)
+2026-10-05T10:18:26 step 15000 train 0.13497 inner-val 0.13919 best 0.13919@15000 (716s)
+2026-10-05T10:19:10 step 16000 train 0.13360 inner-val 0.13786 best 0.13786@16000 (759s)
+2026-10-05T10:19:58 step 17000 train 0.13196 inner-val 0.13646 best 0.13646@17000 (808s)
+2026-10-05T10:20:45 step 18000 train 0.13099 inner-val 0.13517 best 0.13517@18000 (854s)
+2026-10-05T10:21:31 step 19000 train 0.12989 inner-val 0.13401 best 0.13401@19000 (901s)
+2026-10-05T10:22:18 step 20000 train 0.12874 inner-val 0.13290 best 0.13290@20000 (948s)
+2026-10-05T10:23:05 step 21000 train 0.12781 inner-val 0.13192 best 0.13192@21000 (994s)
+2026-10-05T10:23:51 step 22000 train 0.12673 inner-val 0.13103 best 0.13103@22000 (1041s)
+2026-10-05T10:24:39 step 23000 train 0.12586 inner-val 0.13019 best 0.13019@23000 (1089s)
+2026-10-05T10:25:26 step 24000 train 0.12525 inner-val 0.12947 best 0.12947@24000 (1136s)
+2026-10-05T10:26:13 step 25000 train 0.12433 inner-val 0.12873 best 0.12873@25000 (1183s)
+2026-10-05T10:26:59 step 26000 train 0.12359 inner-val 0.12798 best 0.12798@26000 (1229s)
+2026-10-05T10:27:46 step 27000 train 0.12282 inner-val 0.12730 best 0.12730@27000 (1276s)
+2026-10-05T10:28:34 step 28000 train 0.12220 inner-val 0.12667 best 0.12667@28000 (1324s)
+2026-10-05T10:29:20 step 29000 train 0.12178 inner-val 0.12610 best 0.12610@29000 (1370s)
+2026-10-05T10:30:07 step 30000 train 0.12111 inner-val 0.12559 best 0.12559@30000 (1416s)
+2026-10-05T10:30:53 step 31000 train 0.12059 inner-val 0.12507 best 0.12507@31000 (1462s)
+2026-10-05T10:31:40 step 32000 train 0.12002 inner-val 0.12458 best 0.12458@32000 (1510s)
+2026-10-05T10:32:27 step 33000 train 0.11991 inner-val 0.12414 best 0.12414@33000 (1556s)
+2026-10-05T10:33:13 step 34000 train 0.11918 inner-val 0.12375 best 0.12375@34000 (1602s)
+2026-10-05T10:34:00 step 35000 train 0.11839 inner-val 0.12332 best 0.12332@35000 (1650s)
+2026-10-05T10:34:48 step 36000 train 0.11830 inner-val 0.12296 best 0.12296@36000 (1697s)
+2026-10-05T10:35:37 step 37000 train 0.11778 inner-val 0.12259 best 0.12259@37000 (1747s)
+2026-10-05T10:36:26 step 38000 train 0.11739 inner-val 0.12223 best 0.12223@38000 (1796s)
+2026-10-05T10:37:16 step 39000 train 0.11724 inner-val 0.12190 best 0.12190@39000 (1845s)
+2026-10-05T10:38:05 step 40000 train 0.11677 inner-val 0.12158 best 0.12158@40000 (1894s)
+2026-10-05T10:38:50 step 41000 train 0.11644 inner-val 0.12128 best 0.12128@41000 (1940s)
+2026-10-05T10:39:36 step 42000 train 0.11637 inner-val 0.12101 best 0.12101@42000 (1986s)
+2026-10-05T10:40:22 step 43000 train 0.11591 inner-val 0.12075 best 0.12075@43000 (2032s)
+2026-10-05T10:41:09 step 44000 train 0.11569 inner-val 0.12050 best 0.12050@44000 (2079s)
+2026-10-05T10:41:55 step 45000 train 0.11536 inner-val 0.12024 best 0.12024@45000 (2124s)
+2026-10-05T10:42:42 step 46000 train 0.11503 inner-val 0.11999 best 0.11999@46000 (2172s)
+2026-10-05T10:43:28 step 47000 train 0.11474 inner-val 0.11977 best 0.11977@47000 (2218s)
+2026-10-05T10:44:16 step 48000 train 0.11439 inner-val 0.11958 best 0.11958@48000 (2266s)
+2026-10-05T10:45:02 step 49000 train 0.11473 inner-val 0.11941 best 0.11941@49000 (2312s)
+2026-10-05T10:45:49 step 50000 train 0.11406 inner-val 0.11925 best 0.11925@50000 (2359s)
+2026-10-05T10:46:36 step 51000 train 0.11390 inner-val 0.11905 best 0.11905@51000 (2406s)
+2026-10-05T10:47:21 step 52000 train 0.11364 inner-val 0.11888 best 0.11888@52000 (2451s)
+2026-10-05T10:48:09 step 53000 train 0.11349 inner-val 0.11871 best 0.11871@53000 (2499s)
+2026-10-05T10:48:57 step 54000 train 0.11339 inner-val 0.11857 best 0.11857@54000 (2546s)
+2026-10-05T10:49:44 step 55000 train 0.11302 inner-val 0.11842 best 0.11842@55000 (2594s)
+2026-10-05T10:50:30 step 56000 train 0.11312 inner-val 0.11823 best 0.11823@56000 (2640s)
+2026-10-05T10:51:18 step 57000 train 0.11268 inner-val 0.11807 best 0.11807@57000 (2688s)
+2026-10-05T10:52:07 step 58000 train 0.11299 inner-val 0.11794 best 0.11794@58000 (2737s)
+2026-10-05T10:52:53 step 59000 train 0.11259 inner-val 0.11783 best 0.11783@59000 (2783s)
+2026-10-05T10:53:40 step 60000 train 0.11238 inner-val 0.11768 best 0.11768@60000 (2830s)
+2026-10-05T10:54:27 step 61000 train 0.11234 inner-val 0.11759 best 0.11759@61000 (2877s)
+2026-10-05T10:55:13 step 62000 train 0.11232 inner-val 0.11751 best 0.11751@62000 (2923s)
+2026-10-05T10:55:58 step 63000 train 0.11196 inner-val 0.11739 best 0.11739@63000 (2968s)
+2026-10-05T10:56:45 step 64000 train 0.11191 inner-val 0.11730 best 0.11730@64000 (3016s)
+2026-10-05T10:57:33 step 65000 train 0.11197 inner-val 0.11723 best 0.11723@65000 (3063s)
+2026-10-05T10:58:21 step 66000 train 0.11176 inner-val 0.11715 best 0.11715@66000 (3111s)
+2026-10-05T10:59:09 step 67000 train 0.11147 inner-val 0.11706 best 0.11706@67000 (3159s)
+2026-10-05T10:59:54 step 68000 train 0.11144 inner-val 0.11698 best 0.11698@68000 (3204s)
+2026-10-05T11:00:41 step 69000 train 0.11138 inner-val 0.11689 best 0.11689@69000 (3251s)
+2026-10-05T11:01:30 step 70000 train 0.11132 inner-val 0.11684 best 0.11684@70000 (3300s)
+2026-10-05T11:02:16 step 71000 train 0.11114 inner-val 0.11676 best 0.11676@71000 (3346s)
+2026-10-05T11:03:02 step 72000 train 0.11103 inner-val 0.11667 best 0.11667@72000 (3392s)
+2026-10-05T11:03:49 step 73000 train 0.11072 inner-val 0.11656 best 0.11656@73000 (3439s)
+2026-10-05T11:04:36 step 74000 train 0.11062 inner-val 0.11647 best 0.11647@74000 (3486s)
+2026-10-05T11:05:20 step 75000 train 0.11075 inner-val 0.11642 best 0.11642@75000 (3530s)
+2026-10-05T11:06:06 step 76000 train 0.11053 inner-val 0.11639 best 0.11639@76000 (3576s)
+2026-10-05T11:06:55 step 77000 train 0.11067 inner-val 0.11635 best 0.11635@77000 (3624s)
+2026-10-05T11:07:43 step 78000 train 0.11054 inner-val 0.11629 best 0.11629@78000 (3673s)
+2026-10-05T11:08:31 step 79000 train 0.11020 inner-val 0.11623 best 0.11623@79000 (3721s)
+2026-10-05T11:09:16 step 80000 train 0.11022 inner-val 0.11618 best 0.11618@80000 (3766s)
+2026-10-05T11:10:01 step 81000 train 0.11030 inner-val 0.11611 best 0.11611@81000 (3810s)
+2026-10-05T11:10:45 step 82000 train 0.11042 inner-val 0.11607 best 0.11607@82000 (3855s)
+2026-10-05T11:11:30 step 83000 train 0.10992 inner-val 0.11603 best 0.11603@83000 (3900s)
+2026-10-05T11:12:19 step 84000 train 0.11008 inner-val 0.11596 best 0.11596@84000 (3949s)
+2026-10-05T11:13:09 step 85000 train 0.10984 inner-val 0.11591 best 0.11591@85000 (3998s)
+2026-10-05T11:13:58 step 86000 train 0.10989 inner-val 0.11590 best 0.11590@86000 (4048s)
+2026-10-05T11:14:47 step 87000 train 0.10989 inner-val 0.11590 best 0.11590@87000 (4097s)
+2026-10-05T11:15:36 step 88000 train 0.10978 inner-val 0.11584 best 0.11584@88000 (4146s)
+2026-10-05T11:16:23 step 89000 train 0.10983 inner-val 0.11579 best 0.11579@89000 (4193s)
+2026-10-05T11:17:10 step 90000 train 0.10968 inner-val 0.11575 best 0.11575@90000 (4240s)
+2026-10-05T11:18:00 step 91000 train 0.10964 inner-val 0.11572 best 0.11572@91000 (4289s)
+2026-10-05T11:18:49 step 92000 train 0.10970 inner-val 0.11571 best 0.11571@92000 (4339s)
+2026-10-05T11:19:39 step 93000 train 0.10961 inner-val 0.11568 best 0.11568@93000 (4388s)
+2026-10-05T11:20:28 step 94000 train 0.10965 inner-val 0.11566 best 0.11566@94000 (4438s)
+2026-10-05T11:21:18 step 95000 train 0.10936 inner-val 0.11565 best 0.11565@95000 (4487s)
+2026-10-05T11:22:07 step 96000 train 0.10956 inner-val 0.11562 best 0.11562@96000 (4537s)
+2026-10-05T11:22:56 step 97000 train 0.10944 inner-val 0.11556 best 0.11556@97000 (4586s)
+2026-10-05T11:23:46 step 98000 train 0.10939 inner-val 0.11551 best 0.11551@98000 (4636s)
+2026-10-05T11:24:36 step 99000 train 0.10913 inner-val 0.11548 best 0.11548@99000 (4686s)
+2026-10-05T11:25:26 step 100000 train 0.10934 inner-val 0.11544 best 0.11544@100000 (4735s)
+2026-10-05T11:25:32 saved /home/sj_server_1/PBISC/immun-pbmc-fm/model_8_sparse_data_diffusion/plans/paired_finetune_intradataset_20260927/runs/ye2026_sepsis/fold0/scratch_full/checkpoint_ft.pt (selected step 100000, inner-val 0.11544 vs pretrained 1.78181); sha de6a5f0feac4