| { | |
| "architectures": [ | |
| "XomdichForConditionalGeneration" | |
| ], | |
| "attention_dropout": 0.1, | |
| "bos_token_id": 1, | |
| "d_ff": 3072, | |
| "d_kv": 192, | |
| "d_model": 768, | |
| "eos_token_id": 2, | |
| "feed_forward_proj": "relu", | |
| "hidden_size": 768, | |
| "max_sequence_length": 512, | |
| "model_type": "hyper-xomdich", | |
| "num_attention_heads": 4, | |
| "num_key_value_heads": 1, | |
| "num_layers": 26, | |
| "pad_token_id": 25000, | |
| "relative_attention_num_buckets": 32, | |
| "torch_dtype": "float32", | |
| "transformers_version": "4.50.3", | |
| "use_cache": true, | |
| "vocab_size": 25001 | |
| } | |