Skip to main content

Kastan's group workspace

Timestamps visible
2022-10-04 21:35:48
❌ Error logging layer norms -- check message: 'DistributedDataParallel' object has no attribute 'embed'
2022-10-04 21:35:48
❌ Error logging layer norms -- check message: 'DistributedDataParallel' object has no attribute 'embed'
2022-10-04 21:35:48
[10/04/22 16:35:47] INFO     colossalai - colossalai - INFO:
2022-10-04 21:35:48
                             /scratch/bbki/kastanday/conda_envs/envs/nice_base/e
2022-10-04 21:35:48
                             nvs/col_ai_quant/lib/python3.9/site-packages/coloss
2022-10-04 21:35:48
                             alai/trainer/hooks/_log_hook.py:99
2022-10-04 21:35:48
                             after_train_epoch
2022-10-04 21:35:48
                    INFO     colossalai - colossalai - INFO: [Epoch 1 / Train]:
2022-10-04 21:35:48
                             Loss = 40.799 | LR = 7.5e-05 | Throughput = 85.956
2022-10-04 21:35:48
                    INFO     colossalai - colossalai - INFO:
2022-10-04 21:35:48
                             /scratch/bbki/kastanday/conda_envs/envs/nice_base/e
2022-10-04 21:35:48
                             nvs/col_ai_quant/lib/python3.9/site-packages/coloss
2022-10-04 21:35:48
                             alai/utils/memory.py:91 report_memory_usage
2022-10-04 21:35:48
                    INFO     colossalai - colossalai - INFO: [Epoch 1 / Train]:
2022-10-04 21:35:48
                             GPU: allocated 1204.87 MB, max allocated 2095.15
2022-10-04 21:35:48
                             MB, cached: 2222.0 MB, max cached: 2222.0 MB
2022-10-04 21:35:48
                    INFO     colossalai - colossalai - INFO:
2022-10-04 21:35:48
                             /u/kastanday/LLM-Distributed-Quantization/wandb_log
2022-10-04 21:35:48
                             s/custom_wandb_log_hook.py:125 after_train
2022-10-04 21:35:48
                    INFO     colossalai - colossalai - INFO: training finished