| { |
| "setting": "KD only", |
| "alpha": 1.0, |
| "order": "math_code", |
| "seed": 0, |
| "exp": 2, |
| "student": "Qwen/Qwen2.5-1.5B-Instruct", |
| "gpu_student": "NVIDIA RTX A6000", |
| "gpu_teachers": "NVIDIA RTX 6000 Ada Generation", |
| "load_h": 0.004196022219128079, |
| "stages": { |
| "stage1_math": { |
| "domain": "math", |
| "steps": 896, |
| "train_h": 3.303278657529089, |
| "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": 896, |
| "math_kl": 0.07530077810959147, |
| "math_ce": 0.8567340597276344, |
| "code_kl": 0.17175962989564256, |
| "code_ce": 0.32558232033350043, |
| "math_hard_kl": 0.08536007468885945, |
| "math_hard_ce": 0.8316712053384876 |
| }, |
| "peak_gb_student": 36.754438400268555, |
| "peak_gb_teachers": 30.664897918701172 |
| }, |
| "stage2_code": { |
| "domain": "code", |
| "steps": 922, |
| "train_h": 3.3554604044887753, |
| "val_start": { |
| "step": 0, |
| "math_kl": 0.07527671583988893, |
| "math_ce": 0.8567280895290307, |
| "code_kl": 0.17173951820648523, |
| "code_ce": 0.3252568803168954, |
| "math_hard_kl": 0.08535065783016751, |
| "math_hard_ce": 0.831437297924297 |
| }, |
| "val_end": { |
| "step": 922, |
| "math_kl": 0.09337877949966755, |
| "math_ce": 0.7638499494598865, |
| "code_kl": 0.07871157455933593, |
| "code_ce": 0.3116804449427862, |
| "math_hard_kl": 0.1183755381662131, |
| "math_hard_ce": 0.7590982600103172 |
| }, |
| "peak_gb_student": 33.701783180236816, |
| "peak_gb_teachers": 30.005428314208984, |
| "bench_h": 0.7047434378332562, |
| "gpu_bench": "NVIDIA RTX A6000" |
| } |
| }, |
| "train_h": 6.658739062017864, |
| "done": true |
| } |
