{
    "architectures": [
        "LlamaNemotronVLForSequenceClassification"
    ],
    "auto_map": {
        "AutoConfig": "configuration_llama_nemotron_vl.LlamaNemotronVLForSequenceClassificationConfig",
        "AutoModel": "modeling_llama_nemotron_vl.LlamaNemotronVLForSequenceClassification",
        "AutoModelForSequenceClassification": "modeling_llama_nemotron_vl.LlamaNemotronVLForSequenceClassification"
    },
    "bidirectional_attention": false,
    "downsample_ratio": 0.5,
    "dtype": "bfloat16",
    "dynamic_image_size": true,
    "force_image_size": 512,
    "has_no_defaults_at_init": true,
    "id2label": {
        "0": "LABEL_0"
    },
    "img_context_token_id": 128258,
    "is_composition": true,
    "keep_aspect_ratio": false,
    "label2id": {
        "LABEL_0": 0
    },
    "llm_config": {
        "_attn_implementation_autoset": true,
        "architectures": [
            "LlamaBidirectionalForSequenceClassification"
        ],
        "attention_bias": false,
        "attention_dropout": 0.0,
        "bos_token_id": 128000,
        "dtype": "bfloat16",
        "eos_token_id": 128001,
        "head_dim": 64,
        "hidden_act": "silu",
        "hidden_size": 2048,
        "id2label": {
            "0": "LABEL_0"
        },
        "initializer_range": 0.02,
        "intermediate_size": 8192,
        "label2id": {
            "LABEL_0": 0
        },
        "max_position_embeddings": 131072,
        "mlp_bias": false,
        "model_type": "llama_bidirec",
        "num_attention_heads": 32,
        "num_hidden_layers": 16,
        "num_key_value_heads": 8,
        "pad_token_id": 128001,
        "pooling": "avg",
        "pretraining_tp": 1,
        "rms_norm_eps": 1e-05,
        "rope_scaling": {
            "factor": 32.0,
            "high_freq_factor": 4.0,
            "low_freq_factor": 1.0,
            "original_max_position_embeddings": 8192,
            "rope_type": "llama3"
        },
        "rope_theta": 500000.0,
        "tie_word_embeddings": true,
        "use_cache": true,
        "vocab_size": 128267
    },
    "loss_version": "efficient_v2_cp_head",
    "max_dynamic_patch": 12,
    "max_input_tiles": 4,
    "min_dynamic_patch": 1,
    "mlp_checkpoint": false,
    "model_type": "llama_nemotron_vl_rerank",
    "modelopt_quantization_config": {
        "component": "text",
        "modelopt_quantized": true,
        "quant_type": "fp8"
    },
    "p_max_length": 10240,
    "pad2square": false,
    "pad_token_id": 128004,
    "passage_prefix": "passage:",
    "pooling": "avg",
    "pre_feature_reduction": false,
    "prompt_template": "v1",
    "ps_version": "v2",
    "q_max_length": 512,
    "query_prefix": "query:",
    "select_layer": -1,
    "template": "bidirectional-llama-retriever",
    "transformers_version": "4.57.6",
    "use_backbone_lora": 0,
    "use_llm_lora": 0,
    "use_thumbnail": true,
    "vision_config": {
        "_attn_implementation_autoset": true,
        "attention_dropout": 0.0,
        "drop_path_rate": 0.1,
        "dtype": "bfloat16",
        "hidden_act": "gelu_pytorch_tanh",
        "hidden_size": 1152,
        "image_size": 512,
        "intermediate_size": 4304,
        "layer_norm_eps": 1e-06,
        "model_type": "siglip_vision_model",
        "num_attention_heads": 16,
        "num_channels": 3,
        "num_hidden_layers": 27,
        "patch_size": 16
    },
    "vocab_size": 128267,
    "quantization_config": {
        "config_groups": {
            "group_0": {
                "input_activations": {
                    "dynamic": false,
                    "num_bits": 8,
                    "type": "float"
                },
                "weights": {
                    "dynamic": false,
                    "num_bits": 8,
                    "type": "float"
                },
                "targets": [
                    "Linear"
                ]
            }
        },
        "ignore": [
            "model.mlp1*",
            "model.vision_model*",
            "*embed_tokens*",
            "score*",
            "model.score*",
            "language_model.score*",
            "model.language_model.score*",
            "classifier*",
            "model.classifier*",
            "language_model.classifier*",
            "model.language_model.classifier*",
            "lm_head*",
            "model.lm_head*",
            "language_model.lm_head*",
            "model.language_model.lm_head*"
        ],
        "quant_algo": "FP8",
        "producer": {
            "name": "modelopt",
            "version": "0.43.0"
        },
        "quant_method": "modelopt"
    }
}