| { |
| "parameter_count": 494272512, |
| "parameter_summary": { |
| "total": 494272512, |
| "parts": { |
| "blocks": 339738624, |
| "lm_head": 77266944, |
| "other": 77266944 |
| } |
| }, |
| "n_layer": 12, |
| "n_head": 12, |
| "n_embd": 1536, |
| "head_dim": 128, |
| "depth_scale_mode": "loop", |
| "core_repetitions": 2, |
| "core_layer_count": 4, |
| "resolved_core_layer_count": 4, |
| "prelude_layer_count": 4, |
| "resolved_prelude_layer_count": 4, |
| "coda_layer_count": 4, |
| "resolved_coda_layer_count": 4, |
| "effective_depth": 16, |
| "expected_effective_depth": 16.0, |
| "recurrence_rmsnorm": "recurrence_only", |
| "recurrence_emb_alpha": 0.707106781187, |
| "decoder_inject": true, |
| "core_residual_branch_scale": 1.0, |
| "resolved_core_residual_branch_scale": null, |
| "core_repetition_schedule": "none", |
| "core_repetition_crossover_fraction": null, |
| "core_repetition_low": 1, |
| "core_repetition_breakpoints": null, |
| "lsched_lr_warmup_steps": 0, |
| "lsched_lr_change_steps": [], |
| "dep_grow_events": [], |
| "target_core_repetitions": 2, |
| "final_eval_active_core_repetitions": 2, |
| "observed_train_core_repetitions_mean": null, |
| "observed_train_core_repetitions_min": null, |
| "observed_train_core_repetitions_max": null, |
| "observed_train_core_repetitions_histogram": null, |
| "recurrence_truncation_schedule": "none", |
| "recurrence_bp_window": 0, |
| "matrix_lr": 0.023644267457764198, |
| "embedding_lr": 0.007566165586484544, |
| "unembedding_lr": 0.0026750434968953593, |
| "lr_multiplier": 1.0, |
| "effective_matrix_lr": 0.023644267457764198, |
| "effective_embedding_lr": 0.007566165586484544, |
| "effective_unembedding_lr": 0.0026750434968953593, |
| "weight_decay": 0.05, |
| "matrix_weight_decay": 0.05, |
| "embedding_weight_decay": 0.05, |
| "unembedding_weight_decay": 0.05, |
| "embedding_wd_skip_norm": 0.0, |
| "adam_betas": [ |
| 0.8, |
| 0.95 |
| ], |
| "adam_eps": 1e-10, |
| "lr_schedule": "wsd", |
| "warmup_steps": 40, |
| "warmdown_ratio": 1.0, |
| "output_multiplier": 1.0, |
| "residual_branch_multiplier": 0.25, |
| "wte_init_std": 0.02, |
| "uniform_init_scale": 0.0441941738242, |
| "total_batch_size": 524288, |
| "device_batch_size": 16, |
| "num_epochs": 1, |
| "estimated_num_iterations": 5709, |
| "lr_schedule_steps": 5656, |
| "resolved_lr_schedule_steps": 5656, |
| "max_train_steps": 5656, |
| "resolved_max_train_steps": 5656, |
| "completed_epochs": 0, |
| "seed": 42, |
| "train_token_limit": 2995026657, |
| "val_token_limit": 10000000, |
| "token_limit_cutoff_devices": 16, |
| "active_train_tokens": 2995026657, |
| "packed_train_sequences_per_epoch": 1461504, |
| "packed_train_tokens_per_epoch": 2993160192, |
| "active_val_packed_sequences": 4864, |
| "active_val_packed_tokens": 9961472, |
| "val_loss": 2.947298953407689, |
| "final_val_loss": 2.947298953407689, |
| "final_train_loss": 2.913729958551019, |
| "best_val_loss": 2.947298953407689, |
| "best_val_bpb": 0.9578928979604535, |
| "best_val_epoch": 1.0, |
| "train_val_per_epoch": [ |
| { |
| "epoch": 1, |
| "step": 5656, |
| "train_loss": 2.913729958551019, |
| "val_loss": 2.947298953407689, |
| "val_bpb": 0.9578928979604535 |
| } |
| ], |
| "core_block_residual_rms": 1.0154887437820435, |
| "last_train_mfu": 46.31069649027914, |
| "last_train_tok_per_sec": 1051800, |
| "last_train_step_time_seconds": 0.49846720695495605, |
| "total_training_time_seconds": 2807.408871650696, |
| "total_wall_time_seconds": 2879.654093027115, |
| "sequence_len": 2048, |
| "window_pattern": "L" |
| } |
|
|
