config.json

{
    "architectures": [
        "Olmo2ForCausalLM"
    ],
    "attention_bias": false,
    "attention_dropout": 0.0,
    "dtype": "bfloat16",
    "eos_token_id": 100257,
    "hidden_act": "silu",
    "hidden_size": 4096,
    "initializer_range": 0.02,
    "intermediate_size": 11008,
    "layer_types": [
        "sliding_attention",
        "sliding_attention",
        "sliding_attention",
        "full_attention",
        "sliding_attention",
        "sliding_attention",
        "sliding_attention",
        "full_attention",
        "sliding_attention",
        "sliding_attention",
        "sliding_attention",
        "full_attention",
        "sliding_attention",
        "sliding_attention",
        "sliding_attention",
        "full_attention",
        "sliding_attention",
        "sliding_attention",
        "sliding_attention",
        "full_attention",
        "sliding_attention",
        "sliding_attention",
        "sliding_attention",
        "full_attention",
        "sliding_attention",
        "sliding_attention",
        "sliding_attention",
        "full_attention",
        "sliding_attention",
        "sliding_attention",
        "sliding_attention",
        "full_attention"
    ],
    "max_position_embeddings": 1048576,
    "model_type": "olmo2",
    "num_attention_heads": 32,
    "num_hidden_layers": 32,
    "num_key_value_heads": 32,
    "pad_token_id": 100277,
    "rms_norm_eps": 1e-06,
    "rope_scaling": {
        "type": "linear",
        "factor": 16.0
    },
    "rope_theta": 50000000,
    "sliding_window": 4096,
    "tie_word_embeddings": false,
    "transformers_version": "4.57.1",
    "use_cache": false,
    "vocab_size": 100278,
    "quantization_config": {
        "config_groups": {
            "group_0": {
                "input_activations": {
                    "dynamic": false,
                    "num_bits": 4,
                    "type": "float",
                    "group_size": 16
                },
                "weights": {
                    "dynamic": false,
                    "num_bits": 4,
                    "type": "float",
                    "group_size": 16
                },
                "targets": [
                    "Linear"
                ]
            }
        },
        "ignore": [
            "lm_head",
            "lm_head"
        ],
        "quant_algo": "NVFP4",
        "producer": {
            "name": "modelopt",
            "version": "0.39.0"
        },
        "quant_method": "modelopt"
    }
}