optimum-intel-internal-testing/tiny-random-trinity

1{2  "architectures": [3    "AfmoeForCausalLM"4  ],5  "attention_dropout": 0.0,6  "auto_map": {7    "AutoConfig": "configuration_afmoe.AfmoeConfig",8    "AutoModel": "modeling_afmoe.AfmoeModel",9    "AutoModelForCausalLM": "modeling_afmoe.AfmoeForCausalLM"10  },11  "dtype": "float32",12  "global_attn_every_n_layers": 4,13  "head_dim": 16,14  "hidden_act": "silu",15  "hidden_size": 32,16  "initializer_range": 0.02,17  "intermediate_size": 6144,18  "layer_types": [19    "sliding_attention",20    "full_attention",21    "sliding_attention",22    "full_attention"23  ],24  "max_position_embeddings": 16384,25  "model_type": "afmoe",26  "moe_intermediate_size": 16,27  "mup_enabled": false,28  "n_group": 1,29  "num_attention_heads": 2,30  "num_dense_layers": 2,31  "num_expert_groups": 1,32  "num_experts": 4,33  "num_experts_per_tok": 2,34  "num_hidden_layers": 4,35  "num_key_value_heads": 2,36  "num_limited_groups": 1,37  "num_shared_experts": 1,38  "rms_norm_eps": 1e-05,39  "rope_scaling": null,40  "rope_theta": 10000.0,41  "route_norm": true,42  "route_scale": 1.0,43  "score_func": "sigmoid",44  "sliding_window": 1024,45  "tie_word_embeddings": false,46  "topk_group": 1,47  "transformers_version": "4.57.3",48  "use_cache": true,49  "vocab_size": 10035250}51