Skip to main content

Amr-amr's group workspace

Timestamps visible
2024-10-31 20:54:45
    throughput/device/tokens_per_second=73,868
2024-10-31 20:54:45
    throughput/device/batches_per_second=2.254
2024-10-31 20:54:46
2024-10-31 16:54:46.409 cn-l005.server.mila.quebec:0 olmo.train:967 INFO [step=111396/262144,epoch=0]
2024-10-31 20:54:46
    optim/total_grad_norm=0.3061
2024-10-31 20:54:46
    train/CrossEntropyLoss=3.029
2024-10-31 20:54:46
    train/Perplexity=20.67
2024-10-31 20:54:46
    throughput/total_tokens=58,403,586,048
2024-10-31 20:54:46
    throughput/total_training_Gflops=14,531,245,796
2024-10-31 20:54:46
    throughput/total_training_log_Gflops=23.40
2024-10-31 20:54:46
    throughput/device/tokens_per_second=73,860
2024-10-31 20:54:46
    throughput/device/batches_per_second=2.254
2024-10-31 20:54:46
2024-10-31 16:54:46.854 cn-l005.server.mila.quebec:0 olmo.train:967 INFO [step=111397/262144,epoch=0]
2024-10-31 20:54:46
    optim/total_grad_norm=0.3257
2024-10-31 20:54:46
    train/CrossEntropyLoss=3.078
2024-10-31 20:54:46
    train/Perplexity=21.72
2024-10-31 20:54:46
    throughput/total_tokens=58,404,110,336
2024-10-31 20:54:46
    throughput/total_training_Gflops=14,531,376,243
2024-10-31 20:54:46
    throughput/total_training_log_Gflops=23.40
2024-10-31 20:54:46
    throughput/device/tokens_per_second=73,851
2024-10-31 20:54:46
    throughput/device/batches_per_second=2.254