Upload logs/p1_master.log with huggingface_hub

logs/p1_master.log

@@ -0,0 +1,11 @@
+[14:00:25] === 680M phase 1: 4k -> 16k (2x H100, cp=2) ===
+[14:00:25] parent: varybatch step 3400, 10.5 tok/param, scores 3.05 (1.06B scores 3.17)
+[14:00:25] data:   Data_Loader/Data/p1mix  b3=50% b2=30% b1=20%   usable 1220 steps, running 1200
+[14:00:25] YaRN:   SCALE=4 ORIG=4096 -> 16384 == CP_T=16384  OK
+[14:00:25] WSD:    warmup 100 stable 1098 tail 2
+[14:00:25] PHASE 1a: 3-step probe (8192 tok/rank)
+[14:00:55] PROBE OK: 3 steps, peak mem: 50117MB of 81559 MiB/card
+[14:00:55] PHASE 1b: training 1200 steps
+[16:39:09] training done, run 12, loss: 2.686266
+[16:39:09]   peak mem mem: 50117MB
+[16:39:09] === PHASE 1 TRAINING COMPLETE ===