{ "architectures": [ "Qwen3NextMemForCausalLM" ], "attention_bias": false, "attention_dropout": 0.0, "bos_token_id": 1, "dtype": "bfloat16", "eos_token_id": 2, "fwpkm_addr_loss": "me", "fwpkm_addr_loss_weight": 1.0, "fwpkm_before_attn": true, "fwpkm_compress_query": null, "fwpkm_compress_value": "zero_mean", "fwpkm_fp32_fw": true, "fwpkm_grad_clip": false, "fwpkm_heads": 1, "fwpkm_k_dim": 512, "fwpkm_layers": [], "fwpkm_loss_type": "mse", "fwpkm_mem_grad_to_values_only": true, "fwpkm_n_subkeys": 512, "fwpkm_optimizer_lr": 1.0, "fwpkm_optimizer_type": "sgd", "fwpkm_optimizer_weight_decay": 0.0, "fwpkm_out_fuse_gate": true, "fwpkm_qk_score_type": "idw", "fwpkm_query_src": "hidden", "fwpkm_score_nonlinear": "softmax", "fwpkm_score_temperature": 1.0, "fwpkm_target_value_lookahead": 1, "fwpkm_topk": 8, "fwpkm_update_chunk_size": 512, "fwpkm_v_dim": 512, "fwpkm_value_src": "hidden", "fwpkm_variant": "pkm", "fwpkm_weight_loss_with_gates": true, "head_dim": 64, "hidden_act": "silu", "hidden_size": 768, "initializer_range": 0.02, "intermediate_size": 2304, "layer_types": [ "full_attention", "full_attention", "full_attention", "full_attention", "full_attention", "full_attention", "full_attention", "full_attention", "full_attention", "full_attention", "full_attention", "full_attention" ], "linear_conv_kernel_dim": 4, "linear_key_head_dim": 64, "linear_num_key_heads": 8, "linear_num_value_heads": 8, "linear_value_head_dim": 64, "max_position_embeddings": 1024, "max_window_layers": 12, "model_type": "qwen3_next_mem", "num_attention_heads": 12, "num_fwpkm_layers": 0, "num_hidden_layers": 12, "num_key_value_heads": 4, "partial_rotary_factor": 0.25, "pkm_heads": 4, "pkm_k_dim": 512, "pkm_layers": [ 2, 6, 10 ], "pkm_n_subkeys": 512, "pkm_query_rmsnorm": true, "pkm_topk": 32, "pkm_v_dim": 512, "rms_norm_eps": 1e-05, "rope_scaling": null, "rope_theta": 10000, "sliding_window": null, "tie_word_embeddings": true, "transformers_version": "4.57.1", "use_cache": true, "use_sliding_window": false, "vocab_size": 32000 }