training_metadata.json

{
  "base_model": "openai/gpt-oss-120b",
  "method": "8GPU_Model_Parallelism_SFTTrainer",
  "quantization": "Mxfp4",
  "model_parallelism": true,
  "gpu_count": 8,
  "lora_config": {
    "r": 16,
    "alpha": 32,
    "target_modules": [
      "q_proj",
      "k_proj",
      "v_proj",
      "o_proj",
      "gate_proj",
      "up_proj",
      "down_proj",
      "router"
    ],
    "optimized_for": "Project_Euler_Mathematical_Reasoning"
  },
  "training_config": {
    "batch_size": 2,
    "gradient_accumulation": 4,
    "effective_batch_size": 8,
    "learning_rate": 0.0001,
    "max_length": 2048,
    "epochs": 1
  },
  "dataset_type": "Project_Euler_Mathematical_Problems",
  "training_date": "2025-08-27T03:37:03.210164",
  "layer_distribution": "36_layers_across_8_H200_GPUs"
}