{
 "setting": "KD only",
 "alpha": 1.0,
 "order": "code_math",
 "seed": 1,
 "exp": 2,
 "student": "Qwen/Qwen2.5-1.5B-Instruct",
 "gpu_student": "NVIDIA RTX A6000",
 "gpu_teachers": "NVIDIA RTX 6000 Ada Generation",
 "load_h": 0.0038557967212465073,
 "stages": {
  "stage1_code": {
   "domain": "code",
   "steps": 922,
   "train_h": 3.483110751774576,
   "val_start": {
    "step": 0,
    "math_kl": 0.15874307828215575,
    "math_ce": 0.6515277156879916,
    "code_kl": 0.14786912634071003,
    "code_ce": 0.30340147782401555,
    "math_hard_kl": 0.24940142917923272,
    "math_hard_ce": 0.6765219455834559
   },
   "val_end": {
    "step": 922,
    "math_kl": 0.16037220667035662,
    "math_ce": 0.6134552778130008,
    "code_kl": 0.07813229218058698,
    "code_ce": 0.3095525097758208,
    "math_hard_kl": 0.24713406450048658,
    "math_hard_ce": 0.6740036130352745
   },
   "peak_gb_student": 33.696606159210205,
   "peak_gb_teachers": 30.005428314208984,
   "bench_h": 0.6688253919945822,
   "gpu_bench": "NVIDIA RTX A6000"
  },
  "stage2_math": {
   "domain": "math",
   "steps": 896,
   "train_h": 2.9952571921216116,
   "val_start": {
    "step": 0,
    "math_kl": 0.16069652883738267,
    "math_ce": 0.6128856295790037,
    "code_kl": 0.07813702286999027,
    "code_ce": 0.3094309619594328,
    "math_hard_kl": 0.2467836188890191,
    "math_hard_ce": 0.6738775405183716
   },
   "val_end": {
    "step": 896,
    "math_kl": 0.07573669431537612,
    "math_ce": 0.8391097003532377,
    "code_kl": 0.10414732518681599,
    "code_ce": 0.3212356887956559,
    "math_hard_kl": 0.08597009330105414,
    "math_hard_ce": 0.8206646253872792
   },
   "peak_gb_student": 36.51224708557129,
   "peak_gb_teachers": 29.517569541931152,
   "bench_h": 0.760836257007387,
   "gpu_bench": "NVIDIA RTX A6000"
  }
 },
 "resumed": {
  "from": "stage1_code",
  "gpu_student": "NVIDIA RTX A6000",
  "gpu_teachers": "NVIDIA RTX A6000",
  "load_h": 0.025301156838734944
 },
 "train_h": 6.478367943896187,
 "done": true
}