| { |
| "architectures": [ |
| "RoseX2ForCausalLM" |
| ], |
| "model_type": "rose_x2", |
| "vocab_size": 32770, |
| "hidden_size": 640, |
| "intermediate_size": 1536, |
| "num_hidden_layers": 30, |
| "num_attention_heads": 10, |
| "num_key_value_heads": 5, |
| "head_dim": 64, |
| "max_position_embeddings": 2048, |
| "hidden_act": "silu", |
| "rms_norm_eps": 1e-06, |
| "tie_word_embeddings": true, |
| "rope_theta": 100000.0, |
| "use_qk_norm": true, |
| "use_xsa": true, |
| "use_diff_attn": true, |
| "use_logit_softcap": true, |
| "logit_softcap_value": 50.0, |
| "refresh_gate_enabled": true, |
| "refresh_gate_inject_layers": [ |
| 9, |
| 19 |
| ], |
| "refresh_gate_kernel_size": 9, |
| "bos_token_id": 0, |
| "eos_token_id": 0, |
| "pad_token_id": 1, |
| "torch_dtype": "float32", |
| "transformers_version": "4.45.0", |
| "use_cache": true |
| } |
