{
  "parameter_count": 1028128768,
  "parameter_summary": {
    "total": 1028128768,
    "parts": {
      "blocks": 822083584,
      "lm_head": 103022592,
      "other": 103022592
    }
  },
  "n_layer": 16,
  "n_head": 16,
  "n_embd": 2048,
  "head_dim": 128,
  "depth_scale_mode": "none",
  "core_repetitions": 1,
  "core_layer_count": null,
  "resolved_core_layer_count": null,
  "prelude_layer_count": null,
  "resolved_prelude_layer_count": null,
  "coda_layer_count": null,
  "resolved_coda_layer_count": null,
  "effective_depth": 16,
  "expected_effective_depth": 16.0,
  "recurrence_rmsnorm": "none",
  "recurrence_emb_alpha": 0.0,
  "decoder_inject": false,
  "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": 1,
  "final_eval_active_core_repetitions": null,
  "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": 0.011044591188628971,
  "embedding_lr": 0.00499821140793984,
  "unembedding_lr": 0.001249552851987721,
  "lr_multiplier": 1.0,
  "effective_matrix_lr": 0.011044591188628971,
  "effective_embedding_lr": 0.00499821140793984,
  "effective_unembedding_lr": 0.001249552851987721,
  "weight_decay": 0.0707106781187,
  "matrix_weight_decay": 0.0707106781187,
  "embedding_weight_decay": 0.0707106781187,
  "unembedding_weight_decay": 0.0707106781187,
  "embedding_wd_skip_norm": 0.0,
  "adam_betas": [
    0.8,
    0.95
  ],
  "adam_eps": 1e-10,
  "lr_schedule": "wsd",
  "warmup_steps": 40,
  "warmdown_ratio": 0.6,
  "output_multiplier": 0.5,
  "residual_branch_multiplier": 0.25,
  "wte_init_std": 0.00707106781187,
  "uniform_init_scale": 0.0625,
  "total_batch_size": 524288,
  "device_batch_size": 8,
  "num_epochs": 1,
  "estimated_num_iterations": 9898,
  "lr_schedule_steps": 9805,
  "resolved_lr_schedule_steps": 9805,
  "max_train_steps": 9805,
  "resolved_max_train_steps": 9805,
  "completed_epochs": 0,
  "seed": 42,
  "train_token_limit": 5192050278,
  "val_token_limit": 10000000,
  "token_limit_cutoff_devices": 32,
  "active_train_tokens": 5192050278,
  "packed_train_sequences_per_epoch": 2533888,
  "packed_train_tokens_per_epoch": 5189402624,
  "active_val_packed_sequences": 4864,
  "active_val_packed_tokens": 9961472,
  "val_loss": 2.825931950619346,
  "final_val_loss": 2.825931950619346,
  "final_train_loss": 2.8559923413901873,
  "best_val_loss": 2.825931950619346,
  "best_val_bpb": 0.9184563731298452,
  "best_val_epoch": 1.0,
  "train_val_per_epoch": [
    {
      "epoch": 1,
      "step": 9805,
      "train_loss": 2.8559923413901873,
      "val_loss": 2.825931950619346,
      "val_bpb": 0.9184563731298452
    }
  ],
  "core_block_residual_rms": 2.283571243286133,
  "last_train_mfu": 49.20680184555938,
  "last_train_tok_per_sec": 653943,
  "last_train_step_time_seconds": 0.8017332553863525,
  "total_training_time_seconds": 7886.047045707703,
  "total_wall_time_seconds": 8016.610172748566,
  "sequence_len": 2048,
  "window_pattern": "L"
}