| { | |
| "_name_or_path": "trained_models/Pretrain_5_1b5_0_0_no_discuss/14", | |
| "architectures": [ | |
| "RwkvForCausalLM" | |
| ], | |
| "attention_hidden_size": 2048, | |
| "bos_token_id": 0, | |
| "context_length": 1024, | |
| "eos_token_id": 0, | |
| "hidden_size": 2048, | |
| "intermediate_size": 8192, | |
| "layer_norm_epsilon": 1e-05, | |
| "model_type": "rwkv", | |
| "num_hidden_layers": 24, | |
| "rescale_every": 6, | |
| "tie_word_embeddings": false, | |
| "torch_dtype": "bfloat16", | |
| "transformers_version": "4.45.2", | |
| "use_cache": true, | |
| "vocab_size": 65536 | |
| } | |