| 2025-11-06 10:17:53 - experiment_save_merged_model - INFO - Starting merged model save process |
| 2025-11-06 10:17:53 - experiment_save_merged_model - INFO - Arguments: {'lambdas_path': '/work/gj26/b20042/LLM-AdaMerge/outputs/deepseek-7b/task-wise/cross_entropy-ep2-10%dataset-lambda06/llm_adamerge_lambdas.json', 'model_config': '/work/gj26/b20042/LLM-AdaMerge/outputs/deepseek-7b/task-wise/cross_entropy-ep2-10%dataset-lambda06/model_config.yaml', 'output_dir': '/work/gj26/b20042/LLM-AdaMerge/mergekit/outputs/deepseek-7b/llmadamerge/task-wise/cross_entropy-ep2-10%dataset/lambda06', 'model_name': 'merged-model', 'push_to_hub': False, 'hub_repo_id': 'lejelly/ds7b-ep2-data10-id4-taskwise-lambda06', 'private': False, 'device': 'cuda', 'debug': False} |
| 2025-11-06 10:17:53 - experiment_save_merged_model - INFO - Loading lambdas from /work/gj26/b20042/LLM-AdaMerge/outputs/deepseek-7b/task-wise/cross_entropy-ep2-10%dataset-lambda06/llm_adamerge_lambdas.json |
| 2025-11-06 10:17:53 - experiment_save_merged_model - INFO - Auto-detected parameter-wise merge from JSON structure |
| 2025-11-06 10:17:53 - experiment_save_merged_model - INFO - Merge type: parameter_wise |
| 2025-11-06 10:17:53 - experiment_save_merged_model - INFO - [Initial] Memory Usage: |
| 2025-11-06 10:17:53 - experiment_save_merged_model - INFO - Process: 0.46 GB (0.2%) |
| 2025-11-06 10:17:53 - experiment_save_merged_model - INFO - System: 8.66 GB / 212.49 GB (8.7%) |
| 2025-11-06 10:17:53 - experiment_save_merged_model - INFO - Available: 194.00 GB |
| 2025-11-06 10:17:53 - experiment_save_merged_model - INFO - GPU 0: Allocated: 0.00 GB, Reserved: 0.00 GB, Total: 94.50 GB |
| 2025-11-06 10:17:53 - experiment_save_merged_model - INFO - Loading models |
| 2025-11-06 10:18:05 - experiment_save_merged_model - INFO - [After loading models] Memory Usage: |
| 2025-11-06 10:18:05 - experiment_save_merged_model - INFO - Process: 0.71 GB (0.3%) |
| 2025-11-06 10:18:05 - experiment_save_merged_model - INFO - System: 50.33 GB / 212.49 GB (31.6%) |
| 2025-11-06 10:18:05 - experiment_save_merged_model - INFO - Available: 145.28 GB |
| 2025-11-06 10:18:05 - experiment_save_merged_model - INFO - GPU 0: Allocated: 38.61 GB, Reserved: 40.64 GB, Total: 94.50 GB |
| 2025-11-06 10:18:05 - experiment_save_merged_model - INFO - Initializing parameter_wise AdaMerge |
| 2025-11-06 10:22:34 - experiment_save_merged_model - INFO - Loading learned lambdas |
| 2025-11-06 10:22:34 - experiment_save_merged_model - INFO - Deleting original models to free memory (task vectors already computed) |
| 2025-11-06 10:22:34 - experiment_save_merged_model - INFO - [Before deleting models] Memory Usage: |
| 2025-11-06 10:22:34 - experiment_save_merged_model - INFO - Process: 39.12 GB (18.4%) |
| 2025-11-06 10:22:34 - experiment_save_merged_model - INFO - System: 103.79 GB / 212.49 GB (56.8%) |
| 2025-11-06 10:22:34 - experiment_save_merged_model - INFO - Available: 91.83 GB |
| 2025-11-06 10:22:34 - experiment_save_merged_model - INFO - GPU 0: Allocated: 38.61 GB, Reserved: 40.64 GB, Total: 94.50 GB |
| 2025-11-06 10:22:34 - experiment_save_merged_model - INFO - Clearing model_loader references |
| 2025-11-06 10:22:34 - experiment_save_merged_model - INFO - Deleting model variables |
| 2025-11-06 10:22:34 - experiment_save_merged_model - INFO - Running garbage collection |
| 2025-11-06 10:22:34 - experiment_save_merged_model - INFO - [After deleting models and GC] Memory Usage: |
| 2025-11-06 10:22:34 - experiment_save_merged_model - INFO - Process: 39.12 GB (18.4%) |
| 2025-11-06 10:22:34 - experiment_save_merged_model - INFO - System: 63.06 GB / 212.49 GB (37.6%) |
| 2025-11-06 10:22:34 - experiment_save_merged_model - INFO - Available: 132.55 GB |
| 2025-11-06 10:22:34 - experiment_save_merged_model - INFO - GPU 0: Allocated: 0.00 GB, Reserved: 0.00 GB, Total: 94.50 GB |
| 2025-11-06 10:22:34 - experiment_save_merged_model - INFO - [After loading lambdas] Memory Usage: |
| 2025-11-06 10:22:34 - experiment_save_merged_model - INFO - Process: 39.12 GB (18.4%) |
| 2025-11-06 10:22:34 - experiment_save_merged_model - INFO - System: 63.06 GB / 212.49 GB (37.6%) |
| 2025-11-06 10:22:34 - experiment_save_merged_model - INFO - Available: 132.55 GB |
| 2025-11-06 10:22:34 - experiment_save_merged_model - INFO - GPU 0: Allocated: 0.00 GB, Reserved: 0.00 GB, Total: 94.50 GB |
| 2025-11-06 10:22:34 - experiment_save_merged_model - INFO - Creating merged model with learned lambdas |
| 2025-11-06 10:22:35 - experiment_save_merged_model - INFO - [After merging models] Memory Usage: |
| 2025-11-06 10:22:35 - experiment_save_merged_model - INFO - Process: 39.07 GB (18.4%) |
| 2025-11-06 10:22:35 - experiment_save_merged_model - INFO - System: 91.97 GB / 212.49 GB (51.2%) |
| 2025-11-06 10:22:35 - experiment_save_merged_model - INFO - Available: 103.64 GB |
| 2025-11-06 10:22:35 - experiment_save_merged_model - INFO - GPU 0: Allocated: 38.61 GB, Reserved: 43.18 GB, Total: 94.50 GB |
| 2025-11-06 10:22:35 - experiment_save_merged_model - INFO - Freeing memory from AdaMerge object (task vectors and base params no longer needed) |
| 2025-11-06 10:22:35 - experiment_save_merged_model - INFO - Deleting task vectors |
| 2025-11-06 10:22:35 - experiment_save_merged_model - INFO - Deleting base params |
| 2025-11-06 10:22:35 - experiment_save_merged_model - INFO - Deleting functional model |
| 2025-11-06 10:22:36 - experiment_save_merged_model - INFO - [After freeing AdaMerge memory] Memory Usage: |
| 2025-11-06 10:22:36 - experiment_save_merged_model - INFO - Process: 0.42 GB (0.2%) |
| 2025-11-06 10:22:36 - experiment_save_merged_model - INFO - System: 50.23 GB / 212.49 GB (31.6%) |
| 2025-11-06 10:22:36 - experiment_save_merged_model - INFO - Available: 145.39 GB |
| 2025-11-06 10:22:36 - experiment_save_merged_model - INFO - GPU 0: Allocated: 38.61 GB, Reserved: 40.49 GB, Total: 94.50 GB |
| 2025-11-06 10:22:36 - experiment_save_merged_model - INFO - Saving merged model to /work/gj26/b20042/LLM-AdaMerge/mergekit/outputs/deepseek-7b/llmadamerge/task-wise/cross_entropy-ep2-10%dataset/lambda06 |
| 2025-11-06 10:22:36 - experiment_save_merged_model - INFO - Moving merged model to CPU for saving |
| 2025-11-06 10:22:36 - experiment_save_merged_model - ERROR - No safetensors files were saved! |
| 2025-11-06 10:22:36 - experiment_save_merged_model - INFO - Attempting alternative save method... |
| 2025-11-06 10:22:36 - experiment_save_merged_model - INFO - Saved as pytorch_model.bin |
| 2025-11-06 10:22:36 - experiment_save_merged_model - INFO - [After saving model] Memory Usage: |
| 2025-11-06 10:22:36 - experiment_save_merged_model - INFO - Process: 0.42 GB (0.2%) |
| 2025-11-06 10:22:36 - experiment_save_merged_model - INFO - System: 50.23 GB / 212.49 GB (31.6%) |
| 2025-11-06 10:22:36 - experiment_save_merged_model - INFO - Available: 145.39 GB |
| 2025-11-06 10:22:36 - experiment_save_merged_model - INFO - GPU 0: Allocated: 38.61 GB, Reserved: 40.49 GB, Total: 94.50 GB |
| 2025-11-06 10:22:36 - experiment_save_merged_model - INFO - Saving tokenizer |
| 2025-11-06 10:22:36 - experiment_save_merged_model - INFO - Copied lambdas file to /work/gj26/b20042/LLM-AdaMerge/mergekit/outputs/deepseek-7b/llmadamerge/task-wise/cross_entropy-ep2-10%dataset/lambda06/learned_lambdas.json |
| 2025-11-06 10:22:36 - experiment_save_merged_model - INFO - Creating model card |
| 2025-11-06 10:22:36 - experiment_save_merged_model - INFO - Cleaning up models |
| 2025-11-06 10:22:36 - experiment_save_merged_model - INFO - [After cleanup] Memory Usage: |
| 2025-11-06 10:22:36 - experiment_save_merged_model - INFO - Process: 0.42 GB (0.2%) |
| 2025-11-06 10:22:36 - experiment_save_merged_model - INFO - System: 50.18 GB / 212.49 GB (31.6%) |
| 2025-11-06 10:22:36 - experiment_save_merged_model - INFO - Available: 145.44 GB |
| 2025-11-06 10:22:36 - experiment_save_merged_model - INFO - GPU 0: Allocated: 38.61 GB, Reserved: 40.49 GB, Total: 94.50 GB |
| 2025-11-06 10:22:36 - experiment_save_merged_model - INFO - Model saved successfully to /work/gj26/b20042/LLM-AdaMerge/mergekit/outputs/deepseek-7b/llmadamerge/task-wise/cross_entropy-ep2-10%dataset/lambda06 |
|
|
