{
  "hub_model_id": "johnbean393/chiboard-1-m0",
  "source_checkpoint": "/workspace/chiboard_m0_h100_full_0709/stage_mixed/checkpoint-14746",
  "completed_at_utc": "2026-07-09T07:57:46Z",
  "base_model": "LiquidAI/LFM2.5-350M-Base",
  "dataset": "johnbean393/chiboard-1-sft",
  "training_layout": "mixed_packed",
  "seq_idx_collator": "enabled",
  "max_steps": 14746,
  "global_step": 14746,
  "num_train_epochs": 1,
  "final_eval": {
    "epoch": 1.0,
    "eval_loss": 0.17373667657375336,
    "eval_mean_token_accuracy": 0.9519697650369391,
    "eval_num_tokens": 3511267149.0,
    "eval_runtime": 10.9865,
    "eval_samples_per_second": 75.547,
    "eval_steps_per_second": 7.555,
    "step": 14746
  },
  "train_runtime": {
    "train_runtime": "3.506e+04",
    "train_samples_per_second": "25.23",
    "train_steps_per_second": "0.421",
    "train_loss": "0.2377",
    "epoch": "1"
  },
  "stage": {
    "name": "mixed",
    "model_source": "LiquidAI/LFM2.5-350M-Base",
    "output_dir": "/workspace/chiboard_m0_h100_full_0709/stage_mixed",
    "train_rows": 54535518,
    "eval_rows": 50000,
    "max_length": 4096,
    "packing": true,
    "per_device_train_batch_size": 10,
    "gradient_accumulation_steps": 6
  },
  "dataset_summary": {
    "training_layout": "mixed_packed",
    "train_rows": 54535518,
    "dev_rows": 50000,
    "mixed_train_shuffle_seed": 20260704,
    "mixed_max_length": 4096,
    "mixed_packing": true
  }
}