Skip to main content

Amr-amr's group workspace

Timestamps visible
2024-11-02 23:30:09
    throughput/device/tokens_per_second=34,726
2024-11-02 23:30:09
    throughput/device/batches_per_second=1.060
2024-11-02 23:30:10
2024-11-02 19:30:10.599 cn-l044.server.mila.quebec:0 olmo.train:967 INFO [step=237468/262144,epoch=0]
2024-11-02 23:30:10
    optim/total_grad_norm=0.2757
2024-11-02 23:30:10
    train/CrossEntropyLoss=2.843
2024-11-02 23:30:10
    train/Perplexity=17.17
2024-11-02 23:30:10
    throughput/total_tokens=124,501,622,784
2024-11-02 23:30:10
    throughput/total_training_Gflops=49,696,490,059
2024-11-02 23:30:10
    throughput/total_training_log_Gflops=24.63
2024-11-02 23:30:10
    throughput/device/tokens_per_second=34,725
2024-11-02 23:30:10
    throughput/device/batches_per_second=1.060
2024-11-02 23:30:11
2024-11-02 19:30:11.542 cn-l044.server.mila.quebec:0 olmo.train:967 INFO [step=237469/262144,epoch=0]
2024-11-02 23:30:11
    optim/total_grad_norm=0.2693
2024-11-02 23:30:11
    train/CrossEntropyLoss=2.903
2024-11-02 23:30:11
    train/Perplexity=18.23
2024-11-02 23:30:11
    throughput/total_tokens=124,502,147,072
2024-11-02 23:30:11
    throughput/total_training_Gflops=49,696,699,336
2024-11-02 23:30:11
    throughput/total_training_log_Gflops=24.63
2024-11-02 23:30:11
    throughput/device/tokens_per_second=34,725
2024-11-02 23:30:11
    throughput/device/batches_per_second=1.060