{ "d_model": 4096, "d_intermediate": 0, "n_layer": 56, "vocab_size": 256000, "ssm_cfg": { "layer": "Mamba2", "d_state": 128, "d_conv": 4, "expand": 2, "headdim": 64, "ngroups": 8, "chunk_size": 128, "rmsnorm": true, "norm_before_gate": false, "use_mem_eff_path": false }, "rms_norm": true, "residual_in_fp32": false, "fused_add_norm": false, "pad_vocab_size_multiple": 128, "tie_embeddings": false }