{ "architectures": [ "AfmoeForCausalLM" ], "attention_bias": false, "attention_dropout": 0.0, "bos_token_id": null, "dtype": "float32", "eos_token_id": null, "global_attn_every_n_layers": 4, "head_dim": 16, "hidden_act": "silu", "hidden_size": 32, "initializer_range": 0.02, "intermediate_size": 6144, "layer_types": [ "sliding_attention", "full_attention", "sliding_attention", "full_attention" ], "max_position_embeddings": 16384, "model_type": "afmoe", "moe_intermediate_size": 16, "mup_enabled": false, "n_group": 1, "num_attention_heads": 2, "num_dense_layers": 2, "num_expert_groups": 1, "num_experts": 4, "num_experts_per_tok": 2, "num_hidden_layers": 4, "num_key_value_heads": 2, "num_limited_groups": 1, "num_shared_experts": 1, "pad_token_id": null, "rms_norm_eps": 1e-05, "rope_parameters": { "rope_theta": 10000.0, "rope_type": "default" }, "rope_theta": 10000.0, "route_norm": true, "route_scale": 1.0, "score_func": "sigmoid", "sliding_window": 1024, "tie_word_embeddings": false, "topk_group": 1, "transformers_version": "5.3.0.dev0", "use_cache": true, "vocab_size": 100352, "transformers.js_config": { "use_external_data_format": { "model.onnx": 1, "model_fp16.onnx": 1 }, "kv_cache_dtype": { "q4f16": "float16", "fp16": "float16" } } }