{
  "architectures": [
    "MolmoForCausalLM"
  ],
  "attention_layer_norm": true,
  "auto_map": {
    "AutoConfig": "config_molmo.MolmoConfig",
    "AutoModelForCausalLM": "modeling_molmo.MolmoForCausalLM"
  },
  "clip_qkv": null,
  "dtype": "float32",
  "embedding_size": 100352,
  "hidden_size": 4096,
  "initializer_range": 0.02,
  "intermediate_size": 22016,
  "layer_norm_eps": 1e-06,
  "layer_norm_type": "rms",
  "max_position_embeddings": 4096,
  "model_type": "molmo",
  "norm_after": true,
  "num_attention_heads": 32,
  "num_hidden_layers": 32,
  "num_key_value_heads": null,
  "qkv_bias": false,
  "quantization_config": {
    "add_skip_keys": false,
    "dequantize_fp32": false,
    "group_size": 0,
    "is_integer": true,
    "modules_dtype_dict": {},
    "modules_to_not_convert": [],
    "non_blocking": false,
    "quant_conv": false,
    "quant_method": "sdnq",
    "quantization_device": null,
    "return_device": null,
    "svd_rank": 32,
    "svd_steps": 8,
    "use_quantized_matmul": false,
    "use_quantized_matmul_conv": false,
    "use_svd": true,
    "weights_dtype": "uint4"
  },
  "rope_theta": 500000.0,
  "tie_word_embeddings": false,
  "transformers_version": "4.57.1",
  "use_cache": true,
  "use_position_ids": true,
  "vocab_size": 100278,
  "weight_tying": false
}