| { |
| "setting": "KD only", |
| "alpha": 1.0, |
| "order": "math_code", |
| "seed": 1, |
| "exp": 2, |
| "student": "Qwen/Qwen2.5-1.5B-Instruct", |
| "gpu_student": "NVIDIA RTX 6000 Ada Generation", |
| "gpu_teachers": "NVIDIA RTX A6000", |
| "load_h": 0.004078492522239685, |
| "stages": { |
| "stage1_math": { |
| "domain": "math", |
| "steps": 896, |
| "train_h": 2.7449205807844796, |
| "val_start": { |
| "step": 0, |
| "math_kl": 0.15902198004314483, |
| "math_ce": 0.6553564874444496, |
| "code_kl": 0.14797331712898912, |
| "code_ce": 0.3033394576202258, |
| "math_hard_kl": 0.24902600898986357, |
| "math_hard_ce": 0.6769691187959327 |
| }, |
| "val_end": { |
| "step": 896, |
| "math_kl": 0.0756993714180177, |
| "math_ce": 0.858021463785887, |
| "code_kl": 0.1720647986767575, |
| "code_ce": 0.3236746077973602, |
| "math_hard_kl": 0.08558017192474197, |
| "math_hard_ce": 0.8342651866385676 |
| }, |
| "peak_gb_student": 36.57978296279907, |
| "peak_gb_teachers": 30.618666172027588 |
| }, |
| "stage2_code": { |
| "domain": "code", |
| "steps": 922, |
| "train_h": 2.6334087841378317, |
| "val_start": { |
| "step": 0, |
| "math_kl": 0.0757168118165207, |
| "math_ce": 0.858557205345532, |
| "code_kl": 0.17211650957781463, |
| "code_ce": 0.32364794010142417, |
| "math_hard_kl": 0.08564324635022826, |
| "math_hard_ce": 0.8344598311839683 |
| }, |
| "val_end": { |
| "step": 922, |
| "math_kl": 0.0941693826888755, |
| "math_ce": 0.7592589879130468, |
| "code_kl": 0.07876284398144404, |
| "code_ce": 0.3094142575261208, |
| "math_hard_kl": 0.11899033841110994, |
| "math_hard_ce": 0.7571135998629719 |
| }, |
| "peak_gb_student": 33.86883306503296, |
| "peak_gb_teachers": 30.005428314208984, |
| "bench_h": 0.5363972081740698, |
| "gpu_bench": "NVIDIA RTX 6000 Ada Generation" |
| } |
| }, |
| "train_h": 5.378329364922312, |
| "done": true |
| } |
