config.json

{
    "architectures": [
        "MambaForCausalLM"
    ],
    "d_model": 2688,
    "n_layer": 28,
    "vocab_size": 65024,
    "ssm_cfg": {},
    "rms_norm": true,
    "residual_in_fp32": true,
    "fused_add_norm": true,
    "pad_vocab_size_multiple": 8,
    "pad_token_id": 0,
    "bos_token_id": 1,
    "eos_token_id": 1,
    "conv_kernel": 4,
    "d_inner": 5376,
    "expand": 2,
    "hidden_act": "silu",
    "hidden_size": 2688,
    "initializer_range": 0.1,
    "intermediate_size": 5376,
    "layer_norm_epsilon": 1e-05,
    "model_type": "mamba",
    "num_hidden_layers": 28,
    "rescale_prenorm_residual": false,
    "state_size": 16,
    "time_step_floor": 0.0001,
    "time_step_init_scheme": "random",
    "time_step_max": 0.1,
    "time_step_min": 0.001,
    "time_step_rank": 168,
    "time_step_scale": 1.0,
    "torch_dtype": "float32",
    "transformers_version": "4.39.0.dev0",
    "use_bias": false,
    "use_cache": true,
    "use_conv_bias": true,
    "tie_word_embeddings": false
}