Datasets:
| { | |
| "checkpoint": "Scripts/Generation_Model/Models_ckpt/260301_content/BERT/LARGE/best.pt", | |
| "tokenizer_path": "/net/work/wangq/CycAMP/Scripts/Generation_Model/Tokenizer/content_vocab.json", | |
| "test_data_path": "Scripts/Generation_Model/Data/processed/hierarchical/test_content.jsonl", | |
| "config": { | |
| "arch": "bert", | |
| "d_model": 768, | |
| "num_layers": 12, | |
| "max_len": 512, | |
| "dropout": 0.4, | |
| "label_smoothing": 0.0 | |
| }, | |
| "evaluation": { | |
| "device": "cuda", | |
| "batch_size": 32, | |
| "num_samples": 75413, | |
| "timestamp": "2026-03-05T09:47:03.328897" | |
| }, | |
| "metrics": { | |
| "loss": 2.3191633232001205, | |
| "perplexity": 10.167164116094021 | |
| } | |
| } |