Step 1 | loss:2.7997887134552 lr:2e-05 tokens_per_second_per_gpu:58.703458626557335 
Step 2 | loss:1.9635694026947021 lr:2e-05 tokens_per_second_per_gpu:206.65369753685596 
Step 3 | loss:1.9302600622177124 lr:2e-05 tokens_per_second_per_gpu:217.77590086061775 
Step 4 | loss:2.5740036964416504 lr:2e-05 tokens_per_second_per_gpu:220.9926872703538 
Step 5 | loss:2.0926806926727295 lr:2e-05 tokens_per_second_per_gpu:220.38571144272848 
Step 6 | loss:2.3473567962646484 lr:2e-05 tokens_per_second_per_gpu:212.24405759850805 
Step 7 | loss:2.1897239685058594 lr:2e-05 tokens_per_second_per_gpu:215.0095077393116 
Step 8 | loss:1.754281759262085 lr:2e-05 tokens_per_second_per_gpu:209.23794840574274 
Step 9 | loss:2.008831024169922 lr:2e-05 tokens_per_second_per_gpu:184.21359242959744 
Step 10 | loss:2.4670588970184326 lr:2e-05 tokens_per_second_per_gpu:215.28094465841255