export_report.json

{
  "model_name": "google/gemma-4-12b-it",
  "text_stack_path": "model.language_model.layers",
  "num_hidden_layers": 48,
  "intermediate_size": 3840,
  "layer_types": [
    "sliding_attention",
    "sliding_attention",
    "sliding_attention",
    "sliding_attention",
    "sliding_attention",
    "full_attention",
    "sliding_attention",
    "sliding_attention",
    "sliding_attention",
    "sliding_attention",
    "sliding_attention",
    "full_attention",
    "sliding_attention",
    "sliding_attention",
    "sliding_attention",
    "sliding_attention",
    "sliding_attention",
    "full_attention",
    "sliding_attention",
    "sliding_attention",
    "sliding_attention",
    "sliding_attention",
    "sliding_attention",
    "full_attention",
    "sliding_attention",
    "sliding_attention",
    "sliding_attention",
    "sliding_attention",
    "sliding_attention",
    "full_attention",
    "sliding_attention",
    "sliding_attention",
    "sliding_attention",
    "sliding_attention",
    "sliding_attention",
    "full_attention",
    "sliding_attention",
    "sliding_attention",
    "sliding_attention",
    "sliding_attention",
    "sliding_attention",
    "full_attention",
    "sliding_attention",
    "sliding_attention",
    "sliding_attention",
    "sliding_attention",
    "sliding_attention",
    "full_attention"
  ],
  "target_arch": {
    "num_hidden_layers": 48,
    "intermediate_size": 3840,
    "actual_params_B": null,
    "layers_to_remove": 0,
    "width_to_cut": 11520,
    "balance": 0.0,
    "mode": "width_only"
  },
  "gguf_quant": "Q4_K_M",
  "source_architectures": [
    "Gemma4UnifiedForConditionalGeneration"
  ],
  "calibration_cache_file": "/tmp/train_combined.json",
  "calibration_sftp_path": "/var/www/html/models/train_combined.json"
}