1{2 "run_id": "pretrain_enlarged102m_seed01_20261030015017",3 "name": "pretrain_enlarged102m",4 "seed": 1,5 "code_commit": "unknown",6 "config_sha256": "efbc5ac5e13ac94c05ef6d273e2701851307b5066ea5160e171e89ea81bb83ed",7 "started_at": "2026-10-30T01:50:17-0400",8 "host": "m3u96b",9 "python": "3.12.15",10 "torch": "2.15.0",11 "command": "/Users/simon-pierreboucher/wp31/prisme-slm/training/pretrain.py --config configs/experiments/pretrain_enlarged_102m.yaml --device mps",12 "state": "completed",13 "wall_seconds": 8527.03,14 "device": "mps",15 "backend": "mps",16 "precision": "fp32",17 "n_params": 102157056,18 "dataset_manifest_sha256": "d8753bdfa9609fce56a7ec2da7c108666185d021e3f921721e4d2241e1d4fe54",19 "checkpoint_parent": null,20 "finished_at": "2026-10-30T04:12:24-0400",21 "train_tokens": 98304000,22 "estimated_train_flops": 60254683398144000,23 "peak_memory_bytes": 28522004480,24 "final_val_loss": 3.8689021825790406,25 "metrics_file": "experiments/runs/pretrain_enlarged102m_seed01_20261030015017/metrics.jsonl"26}