Upload log_ts_e5.txt with huggingface_hub

log_ts_e5.txt

@@ -0,0 +1,15 @@
+/tmp/tiny_train.py:234: UserWarning: Converting a tensor with requires_grad=True to a scalar may lead to unexpected behavior.
+Consider using tensor.detach() first. (Triggered internally at /pytorch/torch/csrc/autograd/generated/python_variable_methods.cpp:836.)
+  total += float(l) * w
+tch ready
+02:01:41 epoch 0 start
+02:02:05 step 78100 loss=3.4492 tok=1279.6M dt=24
+02:02:20 step 78200 loss=3.1403 tok=1281.2M dt=39
+02:02:34 step 78300 loss=3.2519 tok=1282.9M dt=53
+02:02:49 step 78400 loss=3.2181 tok=1284.5M dt=68
+02:03:03 step 78500 loss=3.4631 tok=1286.1M dt=82
+02:03:18 step 78600 loss=3.4332 tok=1287.8M dt=97
+02:03:33 step 78700 loss=3.1962 tok=1289.4M dt=112
+02:03:48 step 78800 loss=3.2953 tok=1291.1M dt=127
+02:04:02 step 78900 loss=3.1548 tok=1292.7M dt=141
+02:04:17 step 79000 loss=3.2490 tok=1294.3M dt=156