Upload log_ts_f_span.txt with huggingface_hub
Browse files- log_ts_f_span.txt +15 -0
log_ts_f_span.txt
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
/tmp/tiny_train.py:237: UserWarning: Converting a tensor with requires_grad=True to a scalar may lead to unexpected behavior.
|
| 2 |
+
Consider using tensor.detach() first. (Triggered internally at /pytorch/torch/csrc/autograd/generated/python_variable_methods.cpp:836.)
|
| 3 |
+
total += float(l) * w
|
| 4 |
+
d batch ready
|
| 5 |
+
10:25:27 epoch 0 start
|
| 6 |
+
10:25:44 step 78100 loss=2.7669 tok=1279.6M dt=17
|
| 7 |
+
10:25:51 step 78200 loss=2.7653 tok=1281.2M dt=24
|
| 8 |
+
10:25:58 step 78300 loss=2.7037 tok=1282.9M dt=31
|
| 9 |
+
10:26:05 step 78400 loss=2.6348 tok=1284.5M dt=39
|
| 10 |
+
10:26:12 step 78500 loss=2.6299 tok=1286.1M dt=46
|
| 11 |
+
10:26:19 step 78600 loss=2.5841 tok=1287.8M dt=53
|
| 12 |
+
10:26:27 step 78700 loss=2.6414 tok=1289.4M dt=60
|
| 13 |
+
10:26:34 step 78800 loss=2.6474 tok=1291.1M dt=67
|
| 14 |
+
10:26:41 step 78900 loss=2.6496 tok=1292.7M dt=74
|
| 15 |
+
10:26:48 step 79000 loss=2.6420 tok=1294.3M dt=81
|