| { |
| "architectures": [ |
| "Qwen3NextForCausalLM" |
| ], |
| "attention_bias": true, |
| "attention_dropout": 0.0, |
| "attn_output_gate": false, |
| "auto_map": { |
| "AutoConfig": "configuration_qwen3_next.Qwen3NextConfig", |
| "AutoModel": "modeling_qwen3_next.Qwen3NextForCausalLM", |
| "AutoModelForCausalLM": "modeling_qwen3_next.Qwen3NextForCausalLM" |
| }, |
| "bos_token_id": 1, |
| "decoder_sparse_step": 1, |
| "dtype": "float32", |
| "enable_qk_norm": false, |
| "eos_token_id": 2, |
| "full_attention_interval": 0, |
| "head_dim": 64, |
| "hidden_act": "silu", |
| "hidden_size": 1920, |
| "initializer_range": 0.02, |
| "intermediate_size": 4800, |
| "layer_types": ["linear_attention", "linear_attention", "linear_attention", "linear_attention", "linear_attention", "linear_attention", "linear_attention", "linear_attention", "linear_attention", "linear_attention", "linear_attention", "linear_attention", "full_attention", "linear_attention", "linear_attention", "linear_attention", "linear_attention", "linear_attention", "linear_attention", "linear_attention", "full_attention", "full_attention", "linear_attention", "full_attention", "linear_attention", "linear_attention", "linear_attention", "linear_attention", "linear_attention", "linear_attention", "linear_attention", "linear_attention", "linear_attention", "linear_attention", "linear_attention", "linear_attention", "linear_attention", "linear_attention", "linear_attention", "linear_attention", "linear_attention", "linear_attention", "linear_attention", "linear_attention", "linear_attention", "linear_attention", "full_attention", "linear_attention", "full_attention", "full_attention", "linear_attention", "linear_attention", "linear_attention", "linear_attention", "linear_attention", "linear_attention"], |
| "linear_conv_kernel_dim": 4, |
| "linear_key_head_dim": 64, |
| "linear_num_key_heads": 8, |
| "linear_num_value_heads": 32, |
| "linear_value_head_dim": 64, |
| "max_position_embeddings": 32768, |
| "mlp_only_layers": [0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55], |
| "num_experts_per_tok": 2, |
| "num_experts": 0, |
| "model_type": "qwen3_next", |
| "moe_intermediate_size": 0, |
| "norm_topk_prob": true, |
| "num_attention_heads": 30, |
| "num_hidden_layers": 56, |
| "num_key_value_heads": 6, |
| "output_router_logits": false, |
| "partial_rotary_factor": 1.0, |
| "rms_norm_eps": 1e-06, |
| "rope_scaling": null, |
| "rope_theta": 640000.0, |
| "router_aux_loss_coef": 0.001, |
| "router_bias": false, |
| "moe_router_score_function": "softmax", |
| "shared_expert_intermediate_size": 0, |
| "use_shared_expert_gate": true, |
| "tie_word_embeddings": false, |
| "transformers_version": "4.57.1", |
| "use_cache": true, |
| "use_sliding_window": false, |
| "ffn_token_shift": null, |
| "ffn_intermediate_token_shift": null, |
| "attn_token_shift": null, |
| "attn_q_token_shift": null, |
| "attn_k_token_shift": null, |
| "attn_v_token_shift": null, |
| "token_shift_conv_size": 4, |
| "token_shift_conv_init": "default", |
| "attn_position_embedding_type": "rope", |
| "rnn_position_embedding_type": "nope", |
| "attn_logits_scaling": null, |
| "vocab_size": 99000 |
| } |
