{ "adapter_dim": 64, "architectures": [ "modeling_aquarius.AquariusModel" ], "auto_map": { "AutoConfig": "configuration_aquarius.AquariusConfig", "AutoModel": "modeling_aquarius.LlamaModel", "AutoModelForCausalLM": "modeling_aquarius.AquariusModel" }, "bos_token_id": 1, "eos_token_id": 2, "hidden_act": "silu", "hidden_size": 4096, "initializer_range": 0.02, "intermediate_size": 11008, "max_position_embeddings": 4096, "model_type": "aquarius", "moe_dtype": "bfloat16", "moe_scaling": 1, "num_attention_heads": 32, "num_experts": 8, "num_hidden_layers": 32, "num_key_value_heads": 32, "output_router_logits": false, "pretraining_tp": 1, "rms_norm_eps": 1e-05, "rope_scaling": null, "tie_word_embeddings": false, "topk": 2, "torch_dtype": "bfloat16", "transformers_version": "4.34.0", "use_cache": true, "vocab_size": 32000 }