| { |
| "checkpoint_id": "ckpt_20260504_020310_1500_22deff1b_9470fbb7", |
| "created_at": "2026-05-04T02:03:10.542391", |
| "iteration": 1500, |
| "epoch": 0, |
| "train_loss": 0.0, |
| "val_loss": 1.8849022722244262, |
| "learning_rate": 4.1972e-05, |
| "model_config": { |
| "n_layer": 4, |
| "n_head": 4, |
| "n_embd": 256, |
| "vocab_size": 50257, |
| "block_size": 1024, |
| "dropout": 0.1, |
| "bias": true, |
| "initial_connections": 0.1, |
| "connection_growth_rate": 0.05, |
| "max_connections": 1.0 |
| }, |
| "training_config": { |
| "learning_rate": 0.0002, |
| "batch_size": 2, |
| "max_iters": 500, |
| "warmup_iters": 5000, |
| "lr_decay_iters": 50000, |
| "min_lr": 1e-05, |
| "weight_decay": 0.1, |
| "grad_clip": 1.0, |
| "enable_curriculum_learning": true, |
| "enable_introspection": true |
| }, |
| "data_config": { |
| "data_dir": "data/nanecho", |
| "batch_size": 2, |
| "block_size": 1024 |
| }, |
| "metrics": { |
| "val_loss": 1.8849022722244262, |
| "connection_ratio": 0.8000000000000002, |
| "tokens_processed": 3072000, |
| "training_speed_iters_per_sec": 0.08458608730624292 |
| }, |
| "tags": [ |
| "phase_adaptive_mastery", |
| "high_quality", |
| "nanecho", |
| "curriculum", |
| "introspection" |
| ], |
| "parent_checkpoint": null, |
| "notes": "Training checkpoint at iteration 1500 (resumed from iteration 1000) | Phase: adaptive_mastery", |
| "file_size_mb": 249.29106044769287, |
| "quality_score": 230400.45035416176 |
| } |
