| { | |
| "attention_bias": false, | |
| "bos_token_id": 1, | |
| "eos_token_id": 2, | |
| "fuse_cross_entropy": true, | |
| "fuse_norm": true, | |
| "hidden_act": "swish", | |
| "hidden_ratio": 4, | |
| "hidden_size": 4096, | |
| "initializer_range": 0.006, | |
| "intermediate_size": 14336, | |
| "model_type": "top_transformer", | |
| "norm_eps": 1e-06, | |
| "num_heads": 32, | |
| "num_hidden_layers": 30, | |
| "num_kv_heads": 8, | |
| "rope_theta": 10000.0, | |
| "tie_word_embeddings": false, | |
| "use_cache": true, | |
| "window_size": null, | |
| "use_top_loss": true, | |
| "top_window_size": 4096 | |
| } |