| { | |
| "_name_or_path": "Bllossom/llama-3.2-Korean-Bllossom-AICA-5B", | |
| "architectures": [ | |
| "MllamaForCausalLM" | |
| ], | |
| "bos_token_id": 128264, | |
| "cross_attention_layers": [ | |
| 3, | |
| 8, | |
| 13, | |
| 18, | |
| 23, | |
| 28, | |
| 33 | |
| ], | |
| "dropout": 0, | |
| "eos_token_id": 128265, | |
| "hidden_act": "silu", | |
| "hidden_size": 3072, | |
| "initializer_range": 0.02, | |
| "intermediate_size": 8192, | |
| "max_position_embeddings": 131072, | |
| "model_type": "mllama_text_model", | |
| "num_attention_heads": 24, | |
| "num_hidden_layers": 35, | |
| "num_key_value_heads": 8, | |
| "pad_token_id": 128265, | |
| "rms_norm_eps": 1e-05, | |
| "rope_scaling": { | |
| "factor": 32.0, | |
| "high_freq_factor": 4.0, | |
| "low_freq_factor": 1.0, | |
| "original_max_position_embeddings": 8192, | |
| "rope_type": "llama3" | |
| }, | |
| "rope_theta": 500000, | |
| "tie_word_embeddings": false, | |
| "torch_dtype": "float16", | |
| "transformers_version": "4.47.1", | |
| "use_cache": true, | |
| "vocab_size": 128266 | |
| } | |