| { | |
| "exp_id": "2026-04-13-000000-kd-pythia28b-feature-deep", | |
| "exp_dir": "/home/coder/experiments/2026-04-13-000000-kd-pythia28b-feature-deep", | |
| "pool_path": "/home/coder/code/data/pool.txt", | |
| "max_tokens": 10000000, | |
| "teacher_model": "EleutherAI/pythia-2.8b", | |
| "n_layer": 16, | |
| "n_head": 12, | |
| "n_embd": 768, | |
| "seq_len": 1024, | |
| "batch_size": 16, | |
| "epochs": 10, | |
| "lr": 0.0005, | |
| "weight_decay": 0.1, | |
| "warmup_pct": 0.05, | |
| "grad_clip": 1.0, | |
| "grad_accum": 16, | |
| "temperature": 1.5, | |
| "alpha_kd": 0.4, | |
| "alpha_feat": 0.2, | |
| "alpha_ce": 0.4 | |
| } |