autoprogrammer/model-merging

1{2  "_name_or_path": "/data3/jucui/output/meta-llama-Llama-3.2-1B-Instruct-uonlp/CulturaX-de-unsupervised-20241029-201015/model",3  "architectures": [4    "LlamaForCausalLM"5  ],6  "attention_bias": false,7  "attention_dropout": 0.0,8  "bos_token_id": 128000,9  "eos_token_id": [10    128001,11    128008,12    12800913  ],14  "head_dim": 64,15  "hidden_act": "silu",16  "hidden_size": 2048,17  "initializer_range": 0.02,18  "intermediate_size": 8192,19  "max_position_embeddings": 131072,20  "mlp_bias": false,21  "model_type": "llama",22  "num_attention_heads": 32,23  "num_hidden_layers": 16,24  "num_key_value_heads": 8,25  "pad_token_id": 128256,26  "pretraining_tp": 1,27  "rms_norm_eps": 1e-05,28  "rope_scaling": {29    "factor": 32.0,30    "high_freq_factor": 4.0,31    "low_freq_factor": 1.0,32    "original_max_position_embeddings": 8192,33    "rope_type": "llama3"34  },35  "rope_theta": 500000.0,36  "tie_word_embeddings": true,37  "torch_dtype": "float32",38  "transformers_version": "4.46.1",39  "use_cache": true,40  "vocab_size": 12825741}42