Skip to main content

Amr-amr's group workspace

Timestamps visible
2024-10-29 18:31:25
    throughput/device/tokens_per_second=214,366
2024-10-29 18:31:25
    throughput/device/batches_per_second=3.271
2024-10-29 18:31:26
2024-10-29 14:31:26.070 cn-l007.server.mila.quebec:0 olmo.train:967 INFO [step=151182/262144,epoch=0]
2024-10-29 18:31:26
    optim/total_grad_norm=0.4147
2024-10-29 18:31:26
    train/CrossEntropyLoss=3.369
2024-10-29 18:31:26
    train/Perplexity=29.04
2024-10-29 18:31:26
    throughput/total_tokens=79,262,908,416
2024-10-29 18:31:26
    throughput/total_training_Gflops=5,467,289,099
2024-10-29 18:31:26
    throughput/total_training_log_Gflops=22.42
2024-10-29 18:31:26
    throughput/device/tokens_per_second=214,202
2024-10-29 18:31:26
    throughput/device/batches_per_second=3.268
2024-10-29 18:31:26
2024-10-29 14:31:26.374 cn-l007.server.mila.quebec:0 olmo.train:967 INFO [step=151183/262144,epoch=0]
2024-10-29 18:31:26
    optim/total_grad_norm=0.4744
2024-10-29 18:31:26
    train/CrossEntropyLoss=3.396
2024-10-29 18:31:26
    train/Perplexity=29.84
2024-10-29 18:31:26
    throughput/total_tokens=79,263,432,704
2024-10-29 18:31:26
    throughput/total_training_Gflops=5,467,325,262
2024-10-29 18:31:26
    throughput/total_training_log_Gflops=22.42
2024-10-29 18:31:26
    throughput/device/tokens_per_second=214,362
2024-10-29 18:31:26
    throughput/device/batches_per_second=3.271