{ "add_gumbel_noise": true, "architectures": [ "PaTHAttentionForCausalLM" ], "bos_token_id": 1, "elementwise_affine": true, "eos_token_id": 2, "fuse_cross_entropy": true, "fuse_linear_cross_entropy": false, "fuse_norm": true, "fuse_swiglu": true, "gumbel_decay_steps": null, "gumbel_hold_step_ratio": 0.1, "gumbel_hold_steps": null, "gumbel_scale": 1.0, "gumbel_schedule": "linear", "gumbel_zero_step_ratio": 0.5, "hidden_act": "swish", "hidden_ratio": 4, "hidden_size": 1536, "initializer_range": 0.02, "intermediate_size": null, "log_router_stats": true, "model_type": "path_attn", "norm_eps": 1e-06, "num_heads": 24, "num_hidden_layers": 21, "num_kv_heads": 24, "router_bias": false, "router_score": "sigmoid", "router_topk": 1, "router_type": "lin", "tie_word_embeddings": false, "torch_dtype": "float32", "transformers_version": "4.55.4", "use_cache": true, "use_forget_gate": false, "use_head_router": false, "use_l2warp": false, "use_low_rank_w": true, "use_w_shortconv": true, "vocab_size": 32000 }