Step 1 | loss:2.575782299041748 lr:2.0000000000000002e-07 tokens_per_second_per_gpu:242.20615514588104 Step 2 | loss:2.7398736476898193 lr:4.0000000000000003e-07 tokens_per_second_per_gpu:206.1023089570347 Step 3 | loss:2.9978206157684326 lr:6.000000000000001e-07 tokens_per_second_per_gpu:221.3595594187193 Step 4 | loss:2.7159111499786377 lr:8.000000000000001e-07 tokens_per_second_per_gpu:216.67620533325845 Step 5 | loss:2.4694080352783203 lr:1.0000000000000002e-06 tokens_per_second_per_gpu:237.62258381187226 Step 6 | loss:2.42387318611145 lr:1.2000000000000002e-06 tokens_per_second_per_gpu:239.15001745420253 Step 7 | loss:2.4981305599212646 lr:1.4000000000000001e-06 tokens_per_second_per_gpu:222.76815891988795 Step 8 | loss:2.6238675117492676 lr:1.6000000000000001e-06 tokens_per_second_per_gpu:224.34891246994644 Step 9 | loss:2.8125081062316895 lr:1.8000000000000001e-06 tokens_per_second_per_gpu:185.2454162476401 Step 10 | loss:2.553760290145874 lr:2.0000000000000003e-06 tokens_per_second_per_gpu:228.6423423443406 Step 11 | loss:2.7380504608154297 lr:2.2e-06 tokens_per_second_per_gpu:236.86353526582633 Step 12 | loss:2.4070146083831787 lr:2.4000000000000003e-06 tokens_per_second_per_gpu:237.81166375260392 Step 13 | loss:2.661756992340088 lr:2.6e-06 tokens_per_second_per_gpu:229.8780782426217 Step 14 | loss:2.443391799926758 lr:2.8000000000000003e-06 tokens_per_second_per_gpu:234.93454129084736 Step 15 | loss:2.693547487258911 lr:3e-06 tokens_per_second_per_gpu:236.07123654615438 Step 16 | loss:2.5031023025512695 lr:3.2000000000000003e-06 tokens_per_second_per_gpu:229.87349357484342