| { |
| "setting": "KD + ground truth CE", |
| "alpha": 0.5, |
| "order": "code_math", |
| "seed": 1, |
| "exp": 2, |
| "student": "Qwen/Qwen2.5-1.5B-Instruct", |
| "gpu_student": "NVIDIA RTX A6000", |
| "gpu_teachers": "NVIDIA RTX A6000", |
| "load_h": 0.004017585515975952, |
| "stages": { |
| "stage1_code": { |
| "domain": "code", |
| "steps": 922, |
| "train_h": 3.3731059963835612, |
| "val_start": { |
| "step": 0, |
| "math_kl": 0.15910778235068246, |
| "math_ce": 0.6515277156879916, |
| "code_kl": 0.14790875576132695, |
| "code_ce": 0.30340147782401555, |
| "math_hard_kl": 0.24894053959908363, |
| "math_hard_ce": 0.6765219455834559 |
| }, |
| "val_end": { |
| "step": 922, |
| "math_kl": 0.1638128164623604, |
| "math_ce": 0.6383362543090072, |
| "code_kl": 0.08894989687466238, |
| "code_ce": 0.2399126688204941, |
| "math_hard_kl": 0.25499699654851005, |
| "math_hard_ce": 0.6836585082107215 |
| }, |
| "peak_gb_student": 33.696606159210205, |
| "peak_gb_teachers": 29.222092628479004, |
| "bench_h": 0.526390543182691, |
| "gpu_bench": "NVIDIA RTX 6000 Ada Generation" |
| }, |
| "stage2_math": { |
| "domain": "math", |
| "steps": 896, |
| "train_h": 3.326509704987208, |
| "val_start": { |
| "step": 0, |
| "math_kl": 0.16382349345881253, |
| "math_ce": 0.6385867793423897, |
| "code_kl": 0.0890299284561657, |
| "code_ce": 0.2398379567632733, |
| "math_hard_kl": 0.25499628740332914, |
| "math_hard_ce": 0.6836816685083316 |
| }, |
| "val_end": { |
| "step": 896, |
| "math_kl": 0.08503447718578555, |
| "math_ce": 0.4850285649944063, |
| "code_kl": 0.10791840146824236, |
| "code_ce": 0.25932052070065925, |
| "math_hard_kl": 0.09326147659684941, |
| "math_hard_ce": 0.6141027500780367 |
| }, |
| "peak_gb_student": 36.51224708557129, |
| "peak_gb_teachers": 29.517569541931152, |
| "bench_h": 0.784459621641371, |
| "gpu_bench": "NVIDIA RTX A6000" |
| } |
| }, |
| "train_h": 6.699615701370769, |
| "done": true |
| } |
