{ "_name_or_path": "meta-llama/Llama-2-7b-hf", "architectures": [ "SparseLlamaForCausalLM" ], "attention_bias": false, "attention_dropout": 0.0, "auto_map": { "AutoConfig": "ugly_utils.SparseLlamaConfig", "AutoModelForCausalLM": "ugly_utils.SparseLlamaForCausalLM" }, "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": "sparse_llama", "num_attention_heads": 32, "num_hidden_layers": 32, "num_key_value_heads": 32, "pretraining_tp": 1, "rms_norm_eps": 1e-05, "rope_scaling": null, "rope_theta": 10000.0, "thresholds": [ 0.021063182502985, 0.029087254777550697, 0.03309928998351097, 0.04513540118932724, 0.061183542013168335, 0.07723169028759003, 0.08324974030256271, 0.09127381443977356, 0.08726178109645844, 0.09127381443977356, 0.09729187190532684, 0.09528584778308868, 0.0992978885769844, 0.10531593859195709, 0.10732196271419525, 0.11334001272916794, 0.11935807019472122, 0.11735205352306366, 0.12136408686637878, 0.11133399605751038, 0.10531593859195709, 0.10130390524864197, 0.10130390524864197, 0.10330992192029953, 0.10531593859195709, 0.11133399605751038, 0.1153460294008255, 0.12337010353803635, 0.13340020179748535, 0.14744232594966888, 0.1574724167585373, 0.15947842597961426 ], "tie_word_embeddings": false, "torch_dtype": "bfloat16", "transformers_version": "4.36.2", "us_sparse_regularization": false, "use_cache": false, "use_graceful_regularization": false, "use_relu": false, "use_sparse_model": true, "use_sparse_predictor": false, "use_sparse_regularization": false, "vocab_size": 32000 }