| { | |
| "a_low_rank_dim": 64, | |
| "architectures": [ | |
| "RWKV7ForCausalLM" | |
| ], | |
| "attn": null, | |
| "attn_mode": "chunk", | |
| "bos_token_id": 1, | |
| "decay_low_rank_dim": 64, | |
| "dtype": "bfloat16", | |
| "eos_token_id": 2, | |
| "fuse_cross_entropy": true, | |
| "fuse_linear_cross_entropy": false, | |
| "fuse_norm": true, | |
| "gate_low_rank_dim": 128, | |
| "head_dim": 64, | |
| "hidden_act": "sqrelu", | |
| "hidden_ratio": 4, | |
| "hidden_size": 512, | |
| "initializer_range": 0.02, | |
| "intermediate_size": null, | |
| "max_position_embeddings": 4096, | |
| "model_type": "rwkv7", | |
| "norm_bias": true, | |
| "norm_eps": 1e-05, | |
| "norm_first": true, | |
| "num_heads": 8, | |
| "num_hidden_layers": 20, | |
| "pad_token_id": null, | |
| "tie_word_embeddings": false, | |
| "transformers_version": "5.9.0", | |
| "use_cache": true, | |
| "use_l2warp": true, | |
| "v_low_rank_dim": 16, | |
| "value_dim": [ | |
| 512, | |
| 512, | |
| 512, | |
| 512, | |
| 512, | |
| 512, | |
| 512, | |
| 512, | |
| 512, | |
| 512, | |
| 512, | |
| 512, | |
| 512, | |
| 512, | |
| 512, | |
| 512, | |
| 512, | |
| 512, | |
| 512, | |
| 512 | |
| ], | |
| "vocab_size": 32000, | |
| "head_init_design": null | |
| } | |