| { | |
| "architectures": [ | |
| "GeoUniForCausalLM" | |
| ], | |
| "auto_map": { | |
| "AutoConfig": "modeling_geouni.GeoUniConfig", | |
| "AutoModelForCausalLM": "modeling_geouni.GeoUniForCausalLM" | |
| }, | |
| "attention_dropout": 0.0, | |
| "bos_token_id": 151643, | |
| "codebook_size": 8192, | |
| "eos_token_id": 151643, | |
| "hidden_act": "silu", | |
| "hidden_size": 1536, | |
| "initializer_range": 0.02, | |
| "intermediate_size": 8960, | |
| "llm_vocab_size": 151665, | |
| "max_position_embeddings": 131072, | |
| "max_window_layers": 21, | |
| "model_type": "geo-uni", | |
| "num_attention_heads": 12, | |
| "num_hidden_layers": 28, | |
| "num_key_value_heads": 2, | |
| "num_new_special_tokens": 10, | |
| "num_vq_tokens": 256, | |
| "rms_norm_eps": 1e-06, | |
| "rope_theta": 10000, | |
| "sliding_window": 4096, | |
| "tie_word_embeddings": false, | |
| "torch_dtype": "bfloat16", | |
| "transformers_version": "4.49.0", | |
| "use_cache": true, | |
| "use_mrope": false, | |
| "use_sliding_window": false, | |
| "vocab_size": 159867 | |
| } | |