| { | |
| "architectures": ["SPTForCausalLM"], | |
| "model_type": "spt", | |
| "n_layers": 12, | |
| "vocab_size": 97, | |
| "hidden_size": 512, | |
| "n_attn_heads": 16, | |
| "n_kv_heads": 16, | |
| "intermediate_size": 2048, | |
| "max_len": 2048, | |
| "residual": true, | |
| "normalise": true, | |
| "bos_token_id": 95, | |
| "eos_token_id": 95, | |
| "pad_token_id": 95, | |
| "unk_token_id": 96 | |
| } |