Skip to main content

Amr-amr's group workspace

Timestamps visible
2024-10-29 22:43:10
    throughput/device/tokens_per_second=123,024
2024-10-29 22:43:10
    throughput/device/batches_per_second=3.754
2024-10-29 22:43:11
2024-10-29 18:43:11.220 cn-l003.server.mila.quebec:0 olmo.train:967 INFO [step=162832/262144,epoch=0]
2024-10-29 22:43:11
    optim/total_grad_norm=0.3606
2024-10-29 22:43:11
    train/CrossEntropyLoss=3.199
2024-10-29 22:43:11
    train/Perplexity=24.52
2024-10-29 22:43:11
    throughput/total_tokens=85,370,863,616
2024-10-29 22:43:11
    throughput/total_training_Gflops=11,323,307,235
2024-10-29 22:43:11
    throughput/total_training_log_Gflops=23.15
2024-10-29 22:43:11
    throughput/device/tokens_per_second=123,022
2024-10-29 22:43:11
    throughput/device/batches_per_second=3.754
2024-10-29 22:43:11
2024-10-29 18:43:11.485 cn-l003.server.mila.quebec:0 olmo.train:967 INFO [step=162833/262144,epoch=0]
2024-10-29 22:43:11
    optim/total_grad_norm=0.3540
2024-10-29 22:43:11
    train/CrossEntropyLoss=3.125
2024-10-29 22:43:11
    train/Perplexity=22.77
2024-10-29 22:43:11
    throughput/total_tokens=85,371,387,904
2024-10-29 22:43:11
    throughput/total_training_Gflops=11,323,376,775
2024-10-29 22:43:11
    throughput/total_training_log_Gflops=23.15
2024-10-29 22:43:11
    throughput/device/tokens_per_second=123,064
2024-10-29 22:43:11
    throughput/device/batches_per_second=3.756