| { |
| "step": 1679, |
| "val_bpb": 0.33559814257945075, |
| "model_config": { |
| "sequence_len": 8196, |
| "vocab_size": 32768, |
| "n_layer": 12, |
| "n_head": 6, |
| "n_kv_head": 6, |
| "n_embd": 768, |
| "window_pattern": "SSSL" |
| }, |
| "user_config": { |
| "run": "dummy", |
| "device_type": "", |
| "fp8": false, |
| "fp8_recipe": "tensorwise", |
| "depth": 12, |
| "aspect_ratio": 64, |
| "head_dim": 128, |
| "max_seq_len": 8196, |
| "window_pattern": "SSSL", |
| "ordered_data": true, |
| "num_iterations": -1, |
| "target_flops": -1.0, |
| "target_param_data_ratio": 8.0, |
| "device_batch_size": 1, |
| "total_batch_size": 524544, |
| "embedding_lr": 0.3, |
| "unembedding_lr": 0.008, |
| "weight_decay": 0.28, |
| "matrix_lr": 0.02, |
| "scalar_lr": 0.5, |
| "warmup_steps": 40, |
| "warmdown_ratio": 0.65, |
| "final_lr_frac": 0.05, |
| "resume_from_step": -1, |
| "eval_every": 250, |
| "eval_tokens": 81960, |
| "core_metric_every": -1, |
| "core_metric_max_per_task": 500, |
| "sample_every": -1, |
| "save_every": -1, |
| "model_tag": "simple_20261101_curriculum" |
| }, |
| "device_batch_size": 1, |
| "max_seq_len": 8196, |
| "total_batch_size": 524544, |
| "dataloader_state_dict": { |
| "pq_idx": 1, |
| "rg_idx": 179, |
| "epoch": 1 |
| }, |
| "loop_state": { |
| "min_val_bpb": 0.33149468985271474, |
| "smooth_train_loss": 0.6637563544034941, |
| "total_training_time": 27801.5051074028 |
| } |
| } |
