| { | |
| "vocab_size": 32000, | |
| "d_model": 2048, | |
| "num_layers": 20, | |
| "num_heads": 16, | |
| "d_ff": 5120, | |
| "max_seq_len": 1024, | |
| "rope_theta": 500000.0, | |
| "dropout_rate": 0.0, | |
| "use_fp8": false, | |
| "torch_dtype": "bf16", | |
| "architectures": [ | |
| "CodeLLM" | |
| ] | |
| } | |
| { | |
| "vocab_size": 32000, | |
| "d_model": 2048, | |
| "num_layers": 20, | |
| "num_heads": 16, | |
| "d_ff": 5120, | |
| "max_seq_len": 1024, | |
| "rope_theta": 500000.0, | |
| "dropout_rate": 0.0, | |
| "use_fp8": false, | |
| "torch_dtype": "bf16", | |
| "architectures": [ | |
| "CodeLLM" | |
| ] | |
| } | |