Archive 4 checkpoints (1-4)

paper-releases/pbisc-diffusion/updates/20261006T033847Z-bulkonly-lora5-k4000/diffusion/dataset/server_1/model8/plans/bulkonly_lora5_k4000_20261006/runs/imperial_eskd/draw0/lora_k4000/train.log

@@ -0,0 +1,3 @@
+2026-10-06T09:56:03 mode=lora trainable=1,296,224 rows train/inner/test=15/0/142 cells(train)=62,720
+2026-10-06T09:56:03 step 0 inner-val loss nan (pretrained)
+2026-10-06T09:58:03 saved /home/sj_server_1/PBISC/immun-pbmc-fm/model_8_sparse_data_diffusion/plans/bulkonly_lora5_k4000_20261006/runs/imperial_eskd/draw0/lora_k4000/checkpoint_ft.pt (selected step 3000, inner-val nan vs pretrained nan); sha ae1f8e837b0c

paper-releases/pbisc-diffusion/updates/20261006T033847Z-bulkonly-lora5-k4000/diffusion/dataset/server_1/model8/plans/bulkonly_lora5_k4000_20261006/runs/imperial_eskd/draw1/lora_k4000/history.json

@@ -0,0 +1,7 @@
+[
+ {
+  "step": 0,
+  "inner_val_loss": NaN,
+  "train_loss": null
+ }
+]

paper-releases/pbisc-diffusion/updates/20261006T033847Z-bulkonly-lora5-k4000/diffusion/dataset/server_1/model8/plans/bulkonly_lora5_k4000_20261006/runs/imperial_eskd/draw1/lora_k4000/train.log

@@ -0,0 +1,3 @@
+2026-10-06T10:35:07 mode=lora trainable=1,296,224 rows train/inner/test=14/0/143 cells(train)=36,553
+2026-10-06T10:35:07 step 0 inner-val loss nan (pretrained)
+2026-10-06T10:36:59 saved /home/sj_server_1/PBISC/immun-pbmc-fm/model_8_sparse_data_diffusion/plans/bulkonly_lora5_k4000_20261006/runs/imperial_eskd/draw1/lora_k4000/checkpoint_ft.pt (selected step 3000, inner-val nan vs pretrained nan); sha 0f4ccc3c12cd

paper-releases/pbisc-diffusion/updates/20261006T033847Z-bulkonly-lora5-k4000/diffusion/dataset/server_1/model8/plans/bulkonly_lora5_k4000_20261006/runs/imperial_eskd/draw2/lora_k4000/history.json

@@ -0,0 +1,7 @@
+[
+ {
+  "step": 0,
+  "inner_val_loss": NaN,
+  "train_loss": null
+ }
+]