Upload 15 files
Browse files- optimizer.pt +3 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- trainer_state.json +1316 -0
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:08d8b6110f30946c8e09685f87195011acceb0d1ae0c0605ca917a7e14c8913b
|
| 3 |
+
size 31779730
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:38600c788169bd0310cb35337beabfec6cb3a8ba30afe7f852dd7213c819b1bb
|
| 3 |
+
size 14244
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5da1e16c544a148301e93754f9bd45191e21057765d09355f060ebc1b44addc6
|
| 3 |
+
size 1064
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1316 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 2.99949083503055,
|
| 5 |
+
"eval_steps": 100,
|
| 6 |
+
"global_step": 5891,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.025458248472505093,
|
| 13 |
+
"grad_norm": 0.7198461890220642,
|
| 14 |
+
"learning_rate": 1e-05,
|
| 15 |
+
"loss": 0.2893,
|
| 16 |
+
"step": 50
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.05091649694501019,
|
| 20 |
+
"grad_norm": 1.0035477876663208,
|
| 21 |
+
"learning_rate": 1e-05,
|
| 22 |
+
"loss": 0.2938,
|
| 23 |
+
"step": 100
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.05091649694501019,
|
| 27 |
+
"eval_loss": 0.30817005038261414,
|
| 28 |
+
"eval_runtime": 22.8318,
|
| 29 |
+
"eval_samples_per_second": 4.38,
|
| 30 |
+
"eval_steps_per_second": 0.569,
|
| 31 |
+
"step": 100
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.07637474541751528,
|
| 35 |
+
"grad_norm": 1.1024292707443237,
|
| 36 |
+
"learning_rate": 1e-05,
|
| 37 |
+
"loss": 0.3017,
|
| 38 |
+
"step": 150
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.10183299389002037,
|
| 42 |
+
"grad_norm": 0.8950141668319702,
|
| 43 |
+
"learning_rate": 1e-05,
|
| 44 |
+
"loss": 0.2912,
|
| 45 |
+
"step": 200
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.10183299389002037,
|
| 49 |
+
"eval_loss": 0.30671748518943787,
|
| 50 |
+
"eval_runtime": 21.6577,
|
| 51 |
+
"eval_samples_per_second": 4.617,
|
| 52 |
+
"eval_steps_per_second": 0.6,
|
| 53 |
+
"step": 200
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 0.12729124236252545,
|
| 57 |
+
"grad_norm": 0.9789568781852722,
|
| 58 |
+
"learning_rate": 1e-05,
|
| 59 |
+
"loss": 0.2928,
|
| 60 |
+
"step": 250
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 0.15274949083503056,
|
| 64 |
+
"grad_norm": 1.0154632329940796,
|
| 65 |
+
"learning_rate": 1e-05,
|
| 66 |
+
"loss": 0.2839,
|
| 67 |
+
"step": 300
|
| 68 |
+
},
|
| 69 |
+
{
|
| 70 |
+
"epoch": 0.15274949083503056,
|
| 71 |
+
"eval_loss": 0.30376389622688293,
|
| 72 |
+
"eval_runtime": 21.5085,
|
| 73 |
+
"eval_samples_per_second": 4.649,
|
| 74 |
+
"eval_steps_per_second": 0.604,
|
| 75 |
+
"step": 300
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 0.17820773930753564,
|
| 79 |
+
"grad_norm": 0.977684736251831,
|
| 80 |
+
"learning_rate": 1e-05,
|
| 81 |
+
"loss": 0.299,
|
| 82 |
+
"step": 350
|
| 83 |
+
},
|
| 84 |
+
{
|
| 85 |
+
"epoch": 0.20366598778004075,
|
| 86 |
+
"grad_norm": 1.02386474609375,
|
| 87 |
+
"learning_rate": 1e-05,
|
| 88 |
+
"loss": 0.2922,
|
| 89 |
+
"step": 400
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 0.20366598778004075,
|
| 93 |
+
"eval_loss": 0.3035086393356323,
|
| 94 |
+
"eval_runtime": 21.9907,
|
| 95 |
+
"eval_samples_per_second": 4.547,
|
| 96 |
+
"eval_steps_per_second": 0.591,
|
| 97 |
+
"step": 400
|
| 98 |
+
},
|
| 99 |
+
{
|
| 100 |
+
"epoch": 0.22912423625254583,
|
| 101 |
+
"grad_norm": 1.0174798965454102,
|
| 102 |
+
"learning_rate": 1e-05,
|
| 103 |
+
"loss": 0.2955,
|
| 104 |
+
"step": 450
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 0.2545824847250509,
|
| 108 |
+
"grad_norm": 1.0312519073486328,
|
| 109 |
+
"learning_rate": 1e-05,
|
| 110 |
+
"loss": 0.3013,
|
| 111 |
+
"step": 500
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 0.2545824847250509,
|
| 115 |
+
"eval_loss": 0.29991400241851807,
|
| 116 |
+
"eval_runtime": 21.9988,
|
| 117 |
+
"eval_samples_per_second": 4.546,
|
| 118 |
+
"eval_steps_per_second": 0.591,
|
| 119 |
+
"step": 500
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 0.280040733197556,
|
| 123 |
+
"grad_norm": 0.8153128623962402,
|
| 124 |
+
"learning_rate": 1e-05,
|
| 125 |
+
"loss": 0.2902,
|
| 126 |
+
"step": 550
|
| 127 |
+
},
|
| 128 |
+
{
|
| 129 |
+
"epoch": 0.3054989816700611,
|
| 130 |
+
"grad_norm": 0.9280871748924255,
|
| 131 |
+
"learning_rate": 1e-05,
|
| 132 |
+
"loss": 0.2933,
|
| 133 |
+
"step": 600
|
| 134 |
+
},
|
| 135 |
+
{
|
| 136 |
+
"epoch": 0.3054989816700611,
|
| 137 |
+
"eval_loss": 0.29873424768447876,
|
| 138 |
+
"eval_runtime": 21.424,
|
| 139 |
+
"eval_samples_per_second": 4.668,
|
| 140 |
+
"eval_steps_per_second": 0.607,
|
| 141 |
+
"step": 600
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 0.33095723014256617,
|
| 145 |
+
"grad_norm": 1.0311402082443237,
|
| 146 |
+
"learning_rate": 1e-05,
|
| 147 |
+
"loss": 0.2871,
|
| 148 |
+
"step": 650
|
| 149 |
+
},
|
| 150 |
+
{
|
| 151 |
+
"epoch": 0.3564154786150713,
|
| 152 |
+
"grad_norm": 1.1811566352844238,
|
| 153 |
+
"learning_rate": 1e-05,
|
| 154 |
+
"loss": 0.2968,
|
| 155 |
+
"step": 700
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 0.3564154786150713,
|
| 159 |
+
"eval_loss": 0.29955142736434937,
|
| 160 |
+
"eval_runtime": 23.1012,
|
| 161 |
+
"eval_samples_per_second": 4.329,
|
| 162 |
+
"eval_steps_per_second": 0.563,
|
| 163 |
+
"step": 700
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.3818737270875764,
|
| 167 |
+
"grad_norm": 1.2003265619277954,
|
| 168 |
+
"learning_rate": 1e-05,
|
| 169 |
+
"loss": 0.2887,
|
| 170 |
+
"step": 750
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.4073319755600815,
|
| 174 |
+
"grad_norm": 1.488318920135498,
|
| 175 |
+
"learning_rate": 1e-05,
|
| 176 |
+
"loss": 0.2891,
|
| 177 |
+
"step": 800
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.4073319755600815,
|
| 181 |
+
"eval_loss": 0.29836124181747437,
|
| 182 |
+
"eval_runtime": 22.3121,
|
| 183 |
+
"eval_samples_per_second": 4.482,
|
| 184 |
+
"eval_steps_per_second": 0.583,
|
| 185 |
+
"step": 800
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.43279022403258655,
|
| 189 |
+
"grad_norm": 0.858514130115509,
|
| 190 |
+
"learning_rate": 1e-05,
|
| 191 |
+
"loss": 0.2985,
|
| 192 |
+
"step": 850
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.45824847250509165,
|
| 196 |
+
"grad_norm": 0.9991205930709839,
|
| 197 |
+
"learning_rate": 1e-05,
|
| 198 |
+
"loss": 0.2877,
|
| 199 |
+
"step": 900
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.45824847250509165,
|
| 203 |
+
"eval_loss": 0.2956816554069519,
|
| 204 |
+
"eval_runtime": 21.6536,
|
| 205 |
+
"eval_samples_per_second": 4.618,
|
| 206 |
+
"eval_steps_per_second": 0.6,
|
| 207 |
+
"step": 900
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 0.48370672097759676,
|
| 211 |
+
"grad_norm": 1.130011796951294,
|
| 212 |
+
"learning_rate": 1e-05,
|
| 213 |
+
"loss": 0.2831,
|
| 214 |
+
"step": 950
|
| 215 |
+
},
|
| 216 |
+
{
|
| 217 |
+
"epoch": 0.5091649694501018,
|
| 218 |
+
"grad_norm": 0.97832852602005,
|
| 219 |
+
"learning_rate": 1e-05,
|
| 220 |
+
"loss": 0.2844,
|
| 221 |
+
"step": 1000
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 0.5091649694501018,
|
| 225 |
+
"eval_loss": 0.29429855942726135,
|
| 226 |
+
"eval_runtime": 21.7902,
|
| 227 |
+
"eval_samples_per_second": 4.589,
|
| 228 |
+
"eval_steps_per_second": 0.597,
|
| 229 |
+
"step": 1000
|
| 230 |
+
},
|
| 231 |
+
{
|
| 232 |
+
"epoch": 0.5346232179226069,
|
| 233 |
+
"grad_norm": 1.1608392000198364,
|
| 234 |
+
"learning_rate": 1e-05,
|
| 235 |
+
"loss": 0.2802,
|
| 236 |
+
"step": 1050
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 0.560081466395112,
|
| 240 |
+
"grad_norm": 0.9106999635696411,
|
| 241 |
+
"learning_rate": 1e-05,
|
| 242 |
+
"loss": 0.2736,
|
| 243 |
+
"step": 1100
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 0.560081466395112,
|
| 247 |
+
"eval_loss": 0.29138484597206116,
|
| 248 |
+
"eval_runtime": 21.847,
|
| 249 |
+
"eval_samples_per_second": 4.577,
|
| 250 |
+
"eval_steps_per_second": 0.595,
|
| 251 |
+
"step": 1100
|
| 252 |
+
},
|
| 253 |
+
{
|
| 254 |
+
"epoch": 0.5855397148676171,
|
| 255 |
+
"grad_norm": 1.077606201171875,
|
| 256 |
+
"learning_rate": 1e-05,
|
| 257 |
+
"loss": 0.2916,
|
| 258 |
+
"step": 1150
|
| 259 |
+
},
|
| 260 |
+
{
|
| 261 |
+
"epoch": 0.6109979633401222,
|
| 262 |
+
"grad_norm": 1.078594446182251,
|
| 263 |
+
"learning_rate": 1e-05,
|
| 264 |
+
"loss": 0.2877,
|
| 265 |
+
"step": 1200
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 0.6109979633401222,
|
| 269 |
+
"eval_loss": 0.2930351495742798,
|
| 270 |
+
"eval_runtime": 21.8584,
|
| 271 |
+
"eval_samples_per_second": 4.575,
|
| 272 |
+
"eval_steps_per_second": 0.595,
|
| 273 |
+
"step": 1200
|
| 274 |
+
},
|
| 275 |
+
{
|
| 276 |
+
"epoch": 0.6364562118126272,
|
| 277 |
+
"grad_norm": 1.044995665550232,
|
| 278 |
+
"learning_rate": 1e-05,
|
| 279 |
+
"loss": 0.2852,
|
| 280 |
+
"step": 1250
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"epoch": 0.6619144602851323,
|
| 284 |
+
"grad_norm": 1.0691392421722412,
|
| 285 |
+
"learning_rate": 1e-05,
|
| 286 |
+
"loss": 0.2914,
|
| 287 |
+
"step": 1300
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"epoch": 0.6619144602851323,
|
| 291 |
+
"eval_loss": 0.29031333327293396,
|
| 292 |
+
"eval_runtime": 21.8883,
|
| 293 |
+
"eval_samples_per_second": 4.569,
|
| 294 |
+
"eval_steps_per_second": 0.594,
|
| 295 |
+
"step": 1300
|
| 296 |
+
},
|
| 297 |
+
{
|
| 298 |
+
"epoch": 0.6873727087576375,
|
| 299 |
+
"grad_norm": 1.165562629699707,
|
| 300 |
+
"learning_rate": 1e-05,
|
| 301 |
+
"loss": 0.2854,
|
| 302 |
+
"step": 1350
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 0.7128309572301426,
|
| 306 |
+
"grad_norm": 1.1224968433380127,
|
| 307 |
+
"learning_rate": 1e-05,
|
| 308 |
+
"loss": 0.2846,
|
| 309 |
+
"step": 1400
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 0.7128309572301426,
|
| 313 |
+
"eval_loss": 0.2897338271141052,
|
| 314 |
+
"eval_runtime": 22.173,
|
| 315 |
+
"eval_samples_per_second": 4.51,
|
| 316 |
+
"eval_steps_per_second": 0.586,
|
| 317 |
+
"step": 1400
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.7382892057026477,
|
| 321 |
+
"grad_norm": 0.9951677918434143,
|
| 322 |
+
"learning_rate": 1e-05,
|
| 323 |
+
"loss": 0.2754,
|
| 324 |
+
"step": 1450
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.7637474541751528,
|
| 328 |
+
"grad_norm": 1.116921305656433,
|
| 329 |
+
"learning_rate": 1e-05,
|
| 330 |
+
"loss": 0.2754,
|
| 331 |
+
"step": 1500
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.7637474541751528,
|
| 335 |
+
"eval_loss": 0.2867298424243927,
|
| 336 |
+
"eval_runtime": 21.7838,
|
| 337 |
+
"eval_samples_per_second": 4.591,
|
| 338 |
+
"eval_steps_per_second": 0.597,
|
| 339 |
+
"step": 1500
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.7892057026476579,
|
| 343 |
+
"grad_norm": 1.104265570640564,
|
| 344 |
+
"learning_rate": 1e-05,
|
| 345 |
+
"loss": 0.2824,
|
| 346 |
+
"step": 1550
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.814663951120163,
|
| 350 |
+
"grad_norm": 0.8793336749076843,
|
| 351 |
+
"learning_rate": 1e-05,
|
| 352 |
+
"loss": 0.2901,
|
| 353 |
+
"step": 1600
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.814663951120163,
|
| 357 |
+
"eval_loss": 0.28652000427246094,
|
| 358 |
+
"eval_runtime": 21.8736,
|
| 359 |
+
"eval_samples_per_second": 4.572,
|
| 360 |
+
"eval_steps_per_second": 0.594,
|
| 361 |
+
"step": 1600
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 0.840122199592668,
|
| 365 |
+
"grad_norm": 1.2304877042770386,
|
| 366 |
+
"learning_rate": 1e-05,
|
| 367 |
+
"loss": 0.2718,
|
| 368 |
+
"step": 1650
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 0.8655804480651731,
|
| 372 |
+
"grad_norm": 0.9079441428184509,
|
| 373 |
+
"learning_rate": 1e-05,
|
| 374 |
+
"loss": 0.2787,
|
| 375 |
+
"step": 1700
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 0.8655804480651731,
|
| 379 |
+
"eval_loss": 0.2839984893798828,
|
| 380 |
+
"eval_runtime": 21.7844,
|
| 381 |
+
"eval_samples_per_second": 4.59,
|
| 382 |
+
"eval_steps_per_second": 0.597,
|
| 383 |
+
"step": 1700
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 0.8910386965376782,
|
| 387 |
+
"grad_norm": 1.359052300453186,
|
| 388 |
+
"learning_rate": 1e-05,
|
| 389 |
+
"loss": 0.2703,
|
| 390 |
+
"step": 1750
|
| 391 |
+
},
|
| 392 |
+
{
|
| 393 |
+
"epoch": 0.9164969450101833,
|
| 394 |
+
"grad_norm": 1.0245873928070068,
|
| 395 |
+
"learning_rate": 1e-05,
|
| 396 |
+
"loss": 0.2678,
|
| 397 |
+
"step": 1800
|
| 398 |
+
},
|
| 399 |
+
{
|
| 400 |
+
"epoch": 0.9164969450101833,
|
| 401 |
+
"eval_loss": 0.2825533151626587,
|
| 402 |
+
"eval_runtime": 22.022,
|
| 403 |
+
"eval_samples_per_second": 4.541,
|
| 404 |
+
"eval_steps_per_second": 0.59,
|
| 405 |
+
"step": 1800
|
| 406 |
+
},
|
| 407 |
+
{
|
| 408 |
+
"epoch": 0.9419551934826884,
|
| 409 |
+
"grad_norm": 0.9011121988296509,
|
| 410 |
+
"learning_rate": 1e-05,
|
| 411 |
+
"loss": 0.2747,
|
| 412 |
+
"step": 1850
|
| 413 |
+
},
|
| 414 |
+
{
|
| 415 |
+
"epoch": 0.9674134419551935,
|
| 416 |
+
"grad_norm": 1.006032943725586,
|
| 417 |
+
"learning_rate": 1e-05,
|
| 418 |
+
"loss": 0.2721,
|
| 419 |
+
"step": 1900
|
| 420 |
+
},
|
| 421 |
+
{
|
| 422 |
+
"epoch": 0.9674134419551935,
|
| 423 |
+
"eval_loss": 0.2824758291244507,
|
| 424 |
+
"eval_runtime": 22.1289,
|
| 425 |
+
"eval_samples_per_second": 4.519,
|
| 426 |
+
"eval_steps_per_second": 0.587,
|
| 427 |
+
"step": 1900
|
| 428 |
+
},
|
| 429 |
+
{
|
| 430 |
+
"epoch": 0.9928716904276986,
|
| 431 |
+
"grad_norm": 0.9993594288825989,
|
| 432 |
+
"learning_rate": 1e-05,
|
| 433 |
+
"loss": 0.2784,
|
| 434 |
+
"step": 1950
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 1.0183299389002036,
|
| 438 |
+
"grad_norm": 0.9710284471511841,
|
| 439 |
+
"learning_rate": 1e-05,
|
| 440 |
+
"loss": 0.2713,
|
| 441 |
+
"step": 2000
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 1.0183299389002036,
|
| 445 |
+
"eval_loss": 0.2820639908313751,
|
| 446 |
+
"eval_runtime": 21.7507,
|
| 447 |
+
"eval_samples_per_second": 4.598,
|
| 448 |
+
"eval_steps_per_second": 0.598,
|
| 449 |
+
"step": 2000
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 1.0437881873727088,
|
| 453 |
+
"grad_norm": 1.2046750783920288,
|
| 454 |
+
"learning_rate": 1e-05,
|
| 455 |
+
"loss": 0.2799,
|
| 456 |
+
"step": 2050
|
| 457 |
+
},
|
| 458 |
+
{
|
| 459 |
+
"epoch": 1.0692464358452138,
|
| 460 |
+
"grad_norm": 0.9969730377197266,
|
| 461 |
+
"learning_rate": 1e-05,
|
| 462 |
+
"loss": 0.2842,
|
| 463 |
+
"step": 2100
|
| 464 |
+
},
|
| 465 |
+
{
|
| 466 |
+
"epoch": 1.0692464358452138,
|
| 467 |
+
"eval_loss": 0.28064805269241333,
|
| 468 |
+
"eval_runtime": 21.7423,
|
| 469 |
+
"eval_samples_per_second": 4.599,
|
| 470 |
+
"eval_steps_per_second": 0.598,
|
| 471 |
+
"step": 2100
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 1.094704684317719,
|
| 475 |
+
"grad_norm": 0.9368526339530945,
|
| 476 |
+
"learning_rate": 1e-05,
|
| 477 |
+
"loss": 0.2783,
|
| 478 |
+
"step": 2150
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 1.120162932790224,
|
| 482 |
+
"grad_norm": 1.2995036840438843,
|
| 483 |
+
"learning_rate": 1e-05,
|
| 484 |
+
"loss": 0.2781,
|
| 485 |
+
"step": 2200
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 1.120162932790224,
|
| 489 |
+
"eval_loss": 0.2789928913116455,
|
| 490 |
+
"eval_runtime": 21.6436,
|
| 491 |
+
"eval_samples_per_second": 4.62,
|
| 492 |
+
"eval_steps_per_second": 0.601,
|
| 493 |
+
"step": 2200
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 1.145621181262729,
|
| 497 |
+
"grad_norm": 1.2737852334976196,
|
| 498 |
+
"learning_rate": 1e-05,
|
| 499 |
+
"loss": 0.2731,
|
| 500 |
+
"step": 2250
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 1.1710794297352343,
|
| 504 |
+
"grad_norm": 1.0202410221099854,
|
| 505 |
+
"learning_rate": 1e-05,
|
| 506 |
+
"loss": 0.273,
|
| 507 |
+
"step": 2300
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 1.1710794297352343,
|
| 511 |
+
"eval_loss": 0.27778831124305725,
|
| 512 |
+
"eval_runtime": 21.9156,
|
| 513 |
+
"eval_samples_per_second": 4.563,
|
| 514 |
+
"eval_steps_per_second": 0.593,
|
| 515 |
+
"step": 2300
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 1.1965376782077393,
|
| 519 |
+
"grad_norm": 0.9710997939109802,
|
| 520 |
+
"learning_rate": 1e-05,
|
| 521 |
+
"loss": 0.2578,
|
| 522 |
+
"step": 2350
|
| 523 |
+
},
|
| 524 |
+
{
|
| 525 |
+
"epoch": 1.2219959266802445,
|
| 526 |
+
"grad_norm": 0.86209636926651,
|
| 527 |
+
"learning_rate": 1e-05,
|
| 528 |
+
"loss": 0.2714,
|
| 529 |
+
"step": 2400
|
| 530 |
+
},
|
| 531 |
+
{
|
| 532 |
+
"epoch": 1.2219959266802445,
|
| 533 |
+
"eval_loss": 0.27559801936149597,
|
| 534 |
+
"eval_runtime": 21.9259,
|
| 535 |
+
"eval_samples_per_second": 4.561,
|
| 536 |
+
"eval_steps_per_second": 0.593,
|
| 537 |
+
"step": 2400
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"epoch": 1.2474541751527495,
|
| 541 |
+
"grad_norm": 1.0652376413345337,
|
| 542 |
+
"learning_rate": 1e-05,
|
| 543 |
+
"loss": 0.265,
|
| 544 |
+
"step": 2450
|
| 545 |
+
},
|
| 546 |
+
{
|
| 547 |
+
"epoch": 1.2729124236252547,
|
| 548 |
+
"grad_norm": 1.002944827079773,
|
| 549 |
+
"learning_rate": 1e-05,
|
| 550 |
+
"loss": 0.2536,
|
| 551 |
+
"step": 2500
|
| 552 |
+
},
|
| 553 |
+
{
|
| 554 |
+
"epoch": 1.2729124236252547,
|
| 555 |
+
"eval_loss": 0.2747356593608856,
|
| 556 |
+
"eval_runtime": 21.9091,
|
| 557 |
+
"eval_samples_per_second": 4.564,
|
| 558 |
+
"eval_steps_per_second": 0.593,
|
| 559 |
+
"step": 2500
|
| 560 |
+
},
|
| 561 |
+
{
|
| 562 |
+
"epoch": 1.2983706720977597,
|
| 563 |
+
"grad_norm": 1.0743255615234375,
|
| 564 |
+
"learning_rate": 1e-05,
|
| 565 |
+
"loss": 0.2742,
|
| 566 |
+
"step": 2550
|
| 567 |
+
},
|
| 568 |
+
{
|
| 569 |
+
"epoch": 1.3238289205702647,
|
| 570 |
+
"grad_norm": 1.2174959182739258,
|
| 571 |
+
"learning_rate": 1e-05,
|
| 572 |
+
"loss": 0.268,
|
| 573 |
+
"step": 2600
|
| 574 |
+
},
|
| 575 |
+
{
|
| 576 |
+
"epoch": 1.3238289205702647,
|
| 577 |
+
"eval_loss": 0.27502280473709106,
|
| 578 |
+
"eval_runtime": 21.9371,
|
| 579 |
+
"eval_samples_per_second": 4.558,
|
| 580 |
+
"eval_steps_per_second": 0.593,
|
| 581 |
+
"step": 2600
|
| 582 |
+
},
|
| 583 |
+
{
|
| 584 |
+
"epoch": 1.34928716904277,
|
| 585 |
+
"grad_norm": 1.1314553022384644,
|
| 586 |
+
"learning_rate": 1e-05,
|
| 587 |
+
"loss": 0.26,
|
| 588 |
+
"step": 2650
|
| 589 |
+
},
|
| 590 |
+
{
|
| 591 |
+
"epoch": 1.374745417515275,
|
| 592 |
+
"grad_norm": 1.007804036140442,
|
| 593 |
+
"learning_rate": 1e-05,
|
| 594 |
+
"loss": 0.2776,
|
| 595 |
+
"step": 2700
|
| 596 |
+
},
|
| 597 |
+
{
|
| 598 |
+
"epoch": 1.374745417515275,
|
| 599 |
+
"eval_loss": 0.27191075682640076,
|
| 600 |
+
"eval_runtime": 21.9972,
|
| 601 |
+
"eval_samples_per_second": 4.546,
|
| 602 |
+
"eval_steps_per_second": 0.591,
|
| 603 |
+
"step": 2700
|
| 604 |
+
},
|
| 605 |
+
{
|
| 606 |
+
"epoch": 1.4002036659877801,
|
| 607 |
+
"grad_norm": 1.1400426626205444,
|
| 608 |
+
"learning_rate": 1e-05,
|
| 609 |
+
"loss": 0.2577,
|
| 610 |
+
"step": 2750
|
| 611 |
+
},
|
| 612 |
+
{
|
| 613 |
+
"epoch": 1.4256619144602851,
|
| 614 |
+
"grad_norm": 0.9301505088806152,
|
| 615 |
+
"learning_rate": 1e-05,
|
| 616 |
+
"loss": 0.2726,
|
| 617 |
+
"step": 2800
|
| 618 |
+
},
|
| 619 |
+
{
|
| 620 |
+
"epoch": 1.4256619144602851,
|
| 621 |
+
"eval_loss": 0.2725023925304413,
|
| 622 |
+
"eval_runtime": 21.722,
|
| 623 |
+
"eval_samples_per_second": 4.604,
|
| 624 |
+
"eval_steps_per_second": 0.598,
|
| 625 |
+
"step": 2800
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 1.4511201629327903,
|
| 629 |
+
"grad_norm": 1.142259120941162,
|
| 630 |
+
"learning_rate": 1e-05,
|
| 631 |
+
"loss": 0.2635,
|
| 632 |
+
"step": 2850
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 1.4765784114052953,
|
| 636 |
+
"grad_norm": 1.1009142398834229,
|
| 637 |
+
"learning_rate": 1e-05,
|
| 638 |
+
"loss": 0.2572,
|
| 639 |
+
"step": 2900
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 1.4765784114052953,
|
| 643 |
+
"eval_loss": 0.2723616063594818,
|
| 644 |
+
"eval_runtime": 21.8662,
|
| 645 |
+
"eval_samples_per_second": 4.573,
|
| 646 |
+
"eval_steps_per_second": 0.595,
|
| 647 |
+
"step": 2900
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 1.5020366598778003,
|
| 651 |
+
"grad_norm": 1.2985098361968994,
|
| 652 |
+
"learning_rate": 1e-05,
|
| 653 |
+
"loss": 0.2477,
|
| 654 |
+
"step": 2950
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 1.5274949083503055,
|
| 658 |
+
"grad_norm": 1.2853788137435913,
|
| 659 |
+
"learning_rate": 1e-05,
|
| 660 |
+
"loss": 0.2518,
|
| 661 |
+
"step": 3000
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 1.5274949083503055,
|
| 665 |
+
"eval_loss": 0.2711648643016815,
|
| 666 |
+
"eval_runtime": 22.1089,
|
| 667 |
+
"eval_samples_per_second": 4.523,
|
| 668 |
+
"eval_steps_per_second": 0.588,
|
| 669 |
+
"step": 3000
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 1.5529531568228105,
|
| 673 |
+
"grad_norm": 1.3646196126937866,
|
| 674 |
+
"learning_rate": 1e-05,
|
| 675 |
+
"loss": 0.2545,
|
| 676 |
+
"step": 3050
|
| 677 |
+
},
|
| 678 |
+
{
|
| 679 |
+
"epoch": 1.5784114052953155,
|
| 680 |
+
"grad_norm": 1.0580254793167114,
|
| 681 |
+
"learning_rate": 1e-05,
|
| 682 |
+
"loss": 0.2665,
|
| 683 |
+
"step": 3100
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 1.5784114052953155,
|
| 687 |
+
"eval_loss": 0.27152860164642334,
|
| 688 |
+
"eval_runtime": 21.563,
|
| 689 |
+
"eval_samples_per_second": 4.638,
|
| 690 |
+
"eval_steps_per_second": 0.603,
|
| 691 |
+
"step": 3100
|
| 692 |
+
},
|
| 693 |
+
{
|
| 694 |
+
"epoch": 1.6038696537678208,
|
| 695 |
+
"grad_norm": 1.529466152191162,
|
| 696 |
+
"learning_rate": 1e-05,
|
| 697 |
+
"loss": 0.2589,
|
| 698 |
+
"step": 3150
|
| 699 |
+
},
|
| 700 |
+
{
|
| 701 |
+
"epoch": 1.629327902240326,
|
| 702 |
+
"grad_norm": 1.0657099485397339,
|
| 703 |
+
"learning_rate": 1e-05,
|
| 704 |
+
"loss": 0.2559,
|
| 705 |
+
"step": 3200
|
| 706 |
+
},
|
| 707 |
+
{
|
| 708 |
+
"epoch": 1.629327902240326,
|
| 709 |
+
"eval_loss": 0.2696399390697479,
|
| 710 |
+
"eval_runtime": 21.6487,
|
| 711 |
+
"eval_samples_per_second": 4.619,
|
| 712 |
+
"eval_steps_per_second": 0.6,
|
| 713 |
+
"step": 3200
|
| 714 |
+
},
|
| 715 |
+
{
|
| 716 |
+
"epoch": 1.654786150712831,
|
| 717 |
+
"grad_norm": 1.0233310461044312,
|
| 718 |
+
"learning_rate": 1e-05,
|
| 719 |
+
"loss": 0.2654,
|
| 720 |
+
"step": 3250
|
| 721 |
+
},
|
| 722 |
+
{
|
| 723 |
+
"epoch": 1.680244399185336,
|
| 724 |
+
"grad_norm": 1.151859998703003,
|
| 725 |
+
"learning_rate": 1e-05,
|
| 726 |
+
"loss": 0.2609,
|
| 727 |
+
"step": 3300
|
| 728 |
+
},
|
| 729 |
+
{
|
| 730 |
+
"epoch": 1.680244399185336,
|
| 731 |
+
"eval_loss": 0.26971447467803955,
|
| 732 |
+
"eval_runtime": 21.8279,
|
| 733 |
+
"eval_samples_per_second": 4.581,
|
| 734 |
+
"eval_steps_per_second": 0.596,
|
| 735 |
+
"step": 3300
|
| 736 |
+
},
|
| 737 |
+
{
|
| 738 |
+
"epoch": 1.7057026476578412,
|
| 739 |
+
"grad_norm": 1.1169921159744263,
|
| 740 |
+
"learning_rate": 1e-05,
|
| 741 |
+
"loss": 0.266,
|
| 742 |
+
"step": 3350
|
| 743 |
+
},
|
| 744 |
+
{
|
| 745 |
+
"epoch": 1.7311608961303462,
|
| 746 |
+
"grad_norm": 0.9375786185264587,
|
| 747 |
+
"learning_rate": 1e-05,
|
| 748 |
+
"loss": 0.2603,
|
| 749 |
+
"step": 3400
|
| 750 |
+
},
|
| 751 |
+
{
|
| 752 |
+
"epoch": 1.7311608961303462,
|
| 753 |
+
"eval_loss": 0.26760581135749817,
|
| 754 |
+
"eval_runtime": 21.6655,
|
| 755 |
+
"eval_samples_per_second": 4.616,
|
| 756 |
+
"eval_steps_per_second": 0.6,
|
| 757 |
+
"step": 3400
|
| 758 |
+
},
|
| 759 |
+
{
|
| 760 |
+
"epoch": 1.7566191446028512,
|
| 761 |
+
"grad_norm": 1.0559574365615845,
|
| 762 |
+
"learning_rate": 1e-05,
|
| 763 |
+
"loss": 0.2644,
|
| 764 |
+
"step": 3450
|
| 765 |
+
},
|
| 766 |
+
{
|
| 767 |
+
"epoch": 1.7820773930753564,
|
| 768 |
+
"grad_norm": 1.1326260566711426,
|
| 769 |
+
"learning_rate": 1e-05,
|
| 770 |
+
"loss": 0.2448,
|
| 771 |
+
"step": 3500
|
| 772 |
+
},
|
| 773 |
+
{
|
| 774 |
+
"epoch": 1.7820773930753564,
|
| 775 |
+
"eval_loss": 0.2665054500102997,
|
| 776 |
+
"eval_runtime": 21.6484,
|
| 777 |
+
"eval_samples_per_second": 4.619,
|
| 778 |
+
"eval_steps_per_second": 0.601,
|
| 779 |
+
"step": 3500
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 1.8075356415478616,
|
| 783 |
+
"grad_norm": 1.396811842918396,
|
| 784 |
+
"learning_rate": 1e-05,
|
| 785 |
+
"loss": 0.252,
|
| 786 |
+
"step": 3550
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 1.8329938900203666,
|
| 790 |
+
"grad_norm": 1.6235796213150024,
|
| 791 |
+
"learning_rate": 1e-05,
|
| 792 |
+
"loss": 0.2587,
|
| 793 |
+
"step": 3600
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 1.8329938900203666,
|
| 797 |
+
"eval_loss": 0.2665034234523773,
|
| 798 |
+
"eval_runtime": 21.9855,
|
| 799 |
+
"eval_samples_per_second": 4.548,
|
| 800 |
+
"eval_steps_per_second": 0.591,
|
| 801 |
+
"step": 3600
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 1.8584521384928716,
|
| 805 |
+
"grad_norm": 1.2640048265457153,
|
| 806 |
+
"learning_rate": 1e-05,
|
| 807 |
+
"loss": 0.2525,
|
| 808 |
+
"step": 3650
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 1.8839103869653768,
|
| 812 |
+
"grad_norm": 1.1669272184371948,
|
| 813 |
+
"learning_rate": 1e-05,
|
| 814 |
+
"loss": 0.2534,
|
| 815 |
+
"step": 3700
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 1.8839103869653768,
|
| 819 |
+
"eval_loss": 0.26496145129203796,
|
| 820 |
+
"eval_runtime": 22.0324,
|
| 821 |
+
"eval_samples_per_second": 4.539,
|
| 822 |
+
"eval_steps_per_second": 0.59,
|
| 823 |
+
"step": 3700
|
| 824 |
+
},
|
| 825 |
+
{
|
| 826 |
+
"epoch": 1.9093686354378818,
|
| 827 |
+
"grad_norm": 1.2015262842178345,
|
| 828 |
+
"learning_rate": 1e-05,
|
| 829 |
+
"loss": 0.2532,
|
| 830 |
+
"step": 3750
|
| 831 |
+
},
|
| 832 |
+
{
|
| 833 |
+
"epoch": 1.9348268839103868,
|
| 834 |
+
"grad_norm": 1.0683043003082275,
|
| 835 |
+
"learning_rate": 1e-05,
|
| 836 |
+
"loss": 0.2496,
|
| 837 |
+
"step": 3800
|
| 838 |
+
},
|
| 839 |
+
{
|
| 840 |
+
"epoch": 1.9348268839103868,
|
| 841 |
+
"eval_loss": 0.2648490369319916,
|
| 842 |
+
"eval_runtime": 21.9752,
|
| 843 |
+
"eval_samples_per_second": 4.551,
|
| 844 |
+
"eval_steps_per_second": 0.592,
|
| 845 |
+
"step": 3800
|
| 846 |
+
},
|
| 847 |
+
{
|
| 848 |
+
"epoch": 1.960285132382892,
|
| 849 |
+
"grad_norm": 1.4562475681304932,
|
| 850 |
+
"learning_rate": 1e-05,
|
| 851 |
+
"loss": 0.242,
|
| 852 |
+
"step": 3850
|
| 853 |
+
},
|
| 854 |
+
{
|
| 855 |
+
"epoch": 1.9857433808553973,
|
| 856 |
+
"grad_norm": 0.9929770231246948,
|
| 857 |
+
"learning_rate": 1e-05,
|
| 858 |
+
"loss": 0.2528,
|
| 859 |
+
"step": 3900
|
| 860 |
+
},
|
| 861 |
+
{
|
| 862 |
+
"epoch": 1.9857433808553973,
|
| 863 |
+
"eval_loss": 0.263571560382843,
|
| 864 |
+
"eval_runtime": 21.7216,
|
| 865 |
+
"eval_samples_per_second": 4.604,
|
| 866 |
+
"eval_steps_per_second": 0.598,
|
| 867 |
+
"step": 3900
|
| 868 |
+
},
|
| 869 |
+
{
|
| 870 |
+
"epoch": 2.011201629327902,
|
| 871 |
+
"grad_norm": 1.0417041778564453,
|
| 872 |
+
"learning_rate": 1e-05,
|
| 873 |
+
"loss": 0.2452,
|
| 874 |
+
"step": 3950
|
| 875 |
+
},
|
| 876 |
+
{
|
| 877 |
+
"epoch": 2.0366598778004072,
|
| 878 |
+
"grad_norm": 1.0510022640228271,
|
| 879 |
+
"learning_rate": 1e-05,
|
| 880 |
+
"loss": 0.2652,
|
| 881 |
+
"step": 4000
|
| 882 |
+
},
|
| 883 |
+
{
|
| 884 |
+
"epoch": 2.0366598778004072,
|
| 885 |
+
"eval_loss": 0.262724906206131,
|
| 886 |
+
"eval_runtime": 22.0633,
|
| 887 |
+
"eval_samples_per_second": 4.532,
|
| 888 |
+
"eval_steps_per_second": 0.589,
|
| 889 |
+
"step": 4000
|
| 890 |
+
},
|
| 891 |
+
{
|
| 892 |
+
"epoch": 2.0621181262729125,
|
| 893 |
+
"grad_norm": 1.383092999458313,
|
| 894 |
+
"learning_rate": 1e-05,
|
| 895 |
+
"loss": 0.2369,
|
| 896 |
+
"step": 4050
|
| 897 |
+
},
|
| 898 |
+
{
|
| 899 |
+
"epoch": 2.0875763747454177,
|
| 900 |
+
"grad_norm": 1.3613831996917725,
|
| 901 |
+
"learning_rate": 1e-05,
|
| 902 |
+
"loss": 0.2493,
|
| 903 |
+
"step": 4100
|
| 904 |
+
},
|
| 905 |
+
{
|
| 906 |
+
"epoch": 2.0875763747454177,
|
| 907 |
+
"eval_loss": 0.2625581622123718,
|
| 908 |
+
"eval_runtime": 21.7251,
|
| 909 |
+
"eval_samples_per_second": 4.603,
|
| 910 |
+
"eval_steps_per_second": 0.598,
|
| 911 |
+
"step": 4100
|
| 912 |
+
},
|
| 913 |
+
{
|
| 914 |
+
"epoch": 2.1130346232179225,
|
| 915 |
+
"grad_norm": 1.0267040729522705,
|
| 916 |
+
"learning_rate": 1e-05,
|
| 917 |
+
"loss": 0.2511,
|
| 918 |
+
"step": 4150
|
| 919 |
+
},
|
| 920 |
+
{
|
| 921 |
+
"epoch": 2.1384928716904277,
|
| 922 |
+
"grad_norm": 0.9748584032058716,
|
| 923 |
+
"learning_rate": 1e-05,
|
| 924 |
+
"loss": 0.2383,
|
| 925 |
+
"step": 4200
|
| 926 |
+
},
|
| 927 |
+
{
|
| 928 |
+
"epoch": 2.1384928716904277,
|
| 929 |
+
"eval_loss": 0.26004984974861145,
|
| 930 |
+
"eval_runtime": 21.6797,
|
| 931 |
+
"eval_samples_per_second": 4.613,
|
| 932 |
+
"eval_steps_per_second": 0.6,
|
| 933 |
+
"step": 4200
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 2.163951120162933,
|
| 937 |
+
"grad_norm": 1.3859432935714722,
|
| 938 |
+
"learning_rate": 1e-05,
|
| 939 |
+
"loss": 0.2544,
|
| 940 |
+
"step": 4250
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 2.189409368635438,
|
| 944 |
+
"grad_norm": 1.598718285560608,
|
| 945 |
+
"learning_rate": 1e-05,
|
| 946 |
+
"loss": 0.2527,
|
| 947 |
+
"step": 4300
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 2.189409368635438,
|
| 951 |
+
"eval_loss": 0.25936272740364075,
|
| 952 |
+
"eval_runtime": 21.803,
|
| 953 |
+
"eval_samples_per_second": 4.587,
|
| 954 |
+
"eval_steps_per_second": 0.596,
|
| 955 |
+
"step": 4300
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 2.214867617107943,
|
| 959 |
+
"grad_norm": 0.9896050691604614,
|
| 960 |
+
"learning_rate": 1e-05,
|
| 961 |
+
"loss": 0.2423,
|
| 962 |
+
"step": 4350
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 2.240325865580448,
|
| 966 |
+
"grad_norm": 1.4134578704833984,
|
| 967 |
+
"learning_rate": 1e-05,
|
| 968 |
+
"loss": 0.2446,
|
| 969 |
+
"step": 4400
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 2.240325865580448,
|
| 973 |
+
"eval_loss": 0.2597595751285553,
|
| 974 |
+
"eval_runtime": 21.9078,
|
| 975 |
+
"eval_samples_per_second": 4.565,
|
| 976 |
+
"eval_steps_per_second": 0.593,
|
| 977 |
+
"step": 4400
|
| 978 |
+
},
|
| 979 |
+
{
|
| 980 |
+
"epoch": 2.2657841140529533,
|
| 981 |
+
"grad_norm": 1.1840572357177734,
|
| 982 |
+
"learning_rate": 1e-05,
|
| 983 |
+
"loss": 0.251,
|
| 984 |
+
"step": 4450
|
| 985 |
+
},
|
| 986 |
+
{
|
| 987 |
+
"epoch": 2.291242362525458,
|
| 988 |
+
"grad_norm": 1.3326150178909302,
|
| 989 |
+
"learning_rate": 1e-05,
|
| 990 |
+
"loss": 0.2504,
|
| 991 |
+
"step": 4500
|
| 992 |
+
},
|
| 993 |
+
{
|
| 994 |
+
"epoch": 2.291242362525458,
|
| 995 |
+
"eval_loss": 0.2583908438682556,
|
| 996 |
+
"eval_runtime": 21.7836,
|
| 997 |
+
"eval_samples_per_second": 4.591,
|
| 998 |
+
"eval_steps_per_second": 0.597,
|
| 999 |
+
"step": 4500
|
| 1000 |
+
},
|
| 1001 |
+
{
|
| 1002 |
+
"epoch": 2.3167006109979633,
|
| 1003 |
+
"grad_norm": 1.4150619506835938,
|
| 1004 |
+
"learning_rate": 1e-05,
|
| 1005 |
+
"loss": 0.2481,
|
| 1006 |
+
"step": 4550
|
| 1007 |
+
},
|
| 1008 |
+
{
|
| 1009 |
+
"epoch": 2.3421588594704685,
|
| 1010 |
+
"grad_norm": 1.4056681394577026,
|
| 1011 |
+
"learning_rate": 1e-05,
|
| 1012 |
+
"loss": 0.2474,
|
| 1013 |
+
"step": 4600
|
| 1014 |
+
},
|
| 1015 |
+
{
|
| 1016 |
+
"epoch": 2.3421588594704685,
|
| 1017 |
+
"eval_loss": 0.25896069407463074,
|
| 1018 |
+
"eval_runtime": 21.9165,
|
| 1019 |
+
"eval_samples_per_second": 4.563,
|
| 1020 |
+
"eval_steps_per_second": 0.593,
|
| 1021 |
+
"step": 4600
|
| 1022 |
+
},
|
| 1023 |
+
{
|
| 1024 |
+
"epoch": 2.3676171079429738,
|
| 1025 |
+
"grad_norm": 1.1109027862548828,
|
| 1026 |
+
"learning_rate": 1e-05,
|
| 1027 |
+
"loss": 0.2447,
|
| 1028 |
+
"step": 4650
|
| 1029 |
+
},
|
| 1030 |
+
{
|
| 1031 |
+
"epoch": 2.3930753564154785,
|
| 1032 |
+
"grad_norm": 1.1159225702285767,
|
| 1033 |
+
"learning_rate": 1e-05,
|
| 1034 |
+
"loss": 0.2482,
|
| 1035 |
+
"step": 4700
|
| 1036 |
+
},
|
| 1037 |
+
{
|
| 1038 |
+
"epoch": 2.3930753564154785,
|
| 1039 |
+
"eval_loss": 0.2578243911266327,
|
| 1040 |
+
"eval_runtime": 21.5943,
|
| 1041 |
+
"eval_samples_per_second": 4.631,
|
| 1042 |
+
"eval_steps_per_second": 0.602,
|
| 1043 |
+
"step": 4700
|
| 1044 |
+
},
|
| 1045 |
+
{
|
| 1046 |
+
"epoch": 2.4185336048879837,
|
| 1047 |
+
"grad_norm": 1.1924070119857788,
|
| 1048 |
+
"learning_rate": 1e-05,
|
| 1049 |
+
"loss": 0.2506,
|
| 1050 |
+
"step": 4750
|
| 1051 |
+
},
|
| 1052 |
+
{
|
| 1053 |
+
"epoch": 2.443991853360489,
|
| 1054 |
+
"grad_norm": 1.5512938499450684,
|
| 1055 |
+
"learning_rate": 1e-05,
|
| 1056 |
+
"loss": 0.2405,
|
| 1057 |
+
"step": 4800
|
| 1058 |
+
},
|
| 1059 |
+
{
|
| 1060 |
+
"epoch": 2.443991853360489,
|
| 1061 |
+
"eval_loss": 0.2575813829898834,
|
| 1062 |
+
"eval_runtime": 22.0278,
|
| 1063 |
+
"eval_samples_per_second": 4.54,
|
| 1064 |
+
"eval_steps_per_second": 0.59,
|
| 1065 |
+
"step": 4800
|
| 1066 |
+
},
|
| 1067 |
+
{
|
| 1068 |
+
"epoch": 2.4694501018329937,
|
| 1069 |
+
"grad_norm": 1.0490576028823853,
|
| 1070 |
+
"learning_rate": 1e-05,
|
| 1071 |
+
"loss": 0.2476,
|
| 1072 |
+
"step": 4850
|
| 1073 |
+
},
|
| 1074 |
+
{
|
| 1075 |
+
"epoch": 2.494908350305499,
|
| 1076 |
+
"grad_norm": 1.0198191404342651,
|
| 1077 |
+
"learning_rate": 1e-05,
|
| 1078 |
+
"loss": 0.2399,
|
| 1079 |
+
"step": 4900
|
| 1080 |
+
},
|
| 1081 |
+
{
|
| 1082 |
+
"epoch": 2.494908350305499,
|
| 1083 |
+
"eval_loss": 0.25637081265449524,
|
| 1084 |
+
"eval_runtime": 21.9152,
|
| 1085 |
+
"eval_samples_per_second": 4.563,
|
| 1086 |
+
"eval_steps_per_second": 0.593,
|
| 1087 |
+
"step": 4900
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 2.520366598778004,
|
| 1091 |
+
"grad_norm": 1.122515320777893,
|
| 1092 |
+
"learning_rate": 1e-05,
|
| 1093 |
+
"loss": 0.2501,
|
| 1094 |
+
"step": 4950
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 2.5458248472505094,
|
| 1098 |
+
"grad_norm": 1.082686424255371,
|
| 1099 |
+
"learning_rate": 1e-05,
|
| 1100 |
+
"loss": 0.2443,
|
| 1101 |
+
"step": 5000
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 2.5458248472505094,
|
| 1105 |
+
"eval_loss": 0.25523728132247925,
|
| 1106 |
+
"eval_runtime": 21.0107,
|
| 1107 |
+
"eval_samples_per_second": 4.759,
|
| 1108 |
+
"eval_steps_per_second": 0.619,
|
| 1109 |
+
"step": 5000
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 2.571283095723014,
|
| 1113 |
+
"grad_norm": 1.0135226249694824,
|
| 1114 |
+
"learning_rate": 1e-05,
|
| 1115 |
+
"loss": 0.2414,
|
| 1116 |
+
"step": 5050
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 2.5967413441955194,
|
| 1120 |
+
"grad_norm": 0.9929371476173401,
|
| 1121 |
+
"learning_rate": 1e-05,
|
| 1122 |
+
"loss": 0.248,
|
| 1123 |
+
"step": 5100
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 2.5967413441955194,
|
| 1127 |
+
"eval_loss": 0.2532651722431183,
|
| 1128 |
+
"eval_runtime": 21.3518,
|
| 1129 |
+
"eval_samples_per_second": 4.683,
|
| 1130 |
+
"eval_steps_per_second": 0.609,
|
| 1131 |
+
"step": 5100
|
| 1132 |
+
},
|
| 1133 |
+
{
|
| 1134 |
+
"epoch": 2.6221995926680246,
|
| 1135 |
+
"grad_norm": 1.1128815412521362,
|
| 1136 |
+
"learning_rate": 1e-05,
|
| 1137 |
+
"loss": 0.2375,
|
| 1138 |
+
"step": 5150
|
| 1139 |
+
},
|
| 1140 |
+
{
|
| 1141 |
+
"epoch": 2.6476578411405294,
|
| 1142 |
+
"grad_norm": 1.5581951141357422,
|
| 1143 |
+
"learning_rate": 1e-05,
|
| 1144 |
+
"loss": 0.2432,
|
| 1145 |
+
"step": 5200
|
| 1146 |
+
},
|
| 1147 |
+
{
|
| 1148 |
+
"epoch": 2.6476578411405294,
|
| 1149 |
+
"eval_loss": 0.2521425485610962,
|
| 1150 |
+
"eval_runtime": 21.0088,
|
| 1151 |
+
"eval_samples_per_second": 4.76,
|
| 1152 |
+
"eval_steps_per_second": 0.619,
|
| 1153 |
+
"step": 5200
|
| 1154 |
+
},
|
| 1155 |
+
{
|
| 1156 |
+
"epoch": 2.6731160896130346,
|
| 1157 |
+
"grad_norm": 1.1291751861572266,
|
| 1158 |
+
"learning_rate": 1e-05,
|
| 1159 |
+
"loss": 0.2415,
|
| 1160 |
+
"step": 5250
|
| 1161 |
+
},
|
| 1162 |
+
{
|
| 1163 |
+
"epoch": 2.69857433808554,
|
| 1164 |
+
"grad_norm": 1.139137625694275,
|
| 1165 |
+
"learning_rate": 1e-05,
|
| 1166 |
+
"loss": 0.2398,
|
| 1167 |
+
"step": 5300
|
| 1168 |
+
},
|
| 1169 |
+
{
|
| 1170 |
+
"epoch": 2.69857433808554,
|
| 1171 |
+
"eval_loss": 0.25268828868865967,
|
| 1172 |
+
"eval_runtime": 21.7715,
|
| 1173 |
+
"eval_samples_per_second": 4.593,
|
| 1174 |
+
"eval_steps_per_second": 0.597,
|
| 1175 |
+
"step": 5300
|
| 1176 |
+
},
|
| 1177 |
+
{
|
| 1178 |
+
"epoch": 2.724032586558045,
|
| 1179 |
+
"grad_norm": 1.0960917472839355,
|
| 1180 |
+
"learning_rate": 1e-05,
|
| 1181 |
+
"loss": 0.2397,
|
| 1182 |
+
"step": 5350
|
| 1183 |
+
},
|
| 1184 |
+
{
|
| 1185 |
+
"epoch": 2.74949083503055,
|
| 1186 |
+
"grad_norm": 1.4622306823730469,
|
| 1187 |
+
"learning_rate": 1e-05,
|
| 1188 |
+
"loss": 0.2371,
|
| 1189 |
+
"step": 5400
|
| 1190 |
+
},
|
| 1191 |
+
{
|
| 1192 |
+
"epoch": 2.74949083503055,
|
| 1193 |
+
"eval_loss": 0.25145551562309265,
|
| 1194 |
+
"eval_runtime": 21.3534,
|
| 1195 |
+
"eval_samples_per_second": 4.683,
|
| 1196 |
+
"eval_steps_per_second": 0.609,
|
| 1197 |
+
"step": 5400
|
| 1198 |
+
},
|
| 1199 |
+
{
|
| 1200 |
+
"epoch": 2.774949083503055,
|
| 1201 |
+
"grad_norm": 1.0508596897125244,
|
| 1202 |
+
"learning_rate": 1e-05,
|
| 1203 |
+
"loss": 0.2389,
|
| 1204 |
+
"step": 5450
|
| 1205 |
+
},
|
| 1206 |
+
{
|
| 1207 |
+
"epoch": 2.8004073319755602,
|
| 1208 |
+
"grad_norm": 0.8241857886314392,
|
| 1209 |
+
"learning_rate": 1e-05,
|
| 1210 |
+
"loss": 0.2403,
|
| 1211 |
+
"step": 5500
|
| 1212 |
+
},
|
| 1213 |
+
{
|
| 1214 |
+
"epoch": 2.8004073319755602,
|
| 1215 |
+
"eval_loss": 0.25168588757514954,
|
| 1216 |
+
"eval_runtime": 21.2395,
|
| 1217 |
+
"eval_samples_per_second": 4.708,
|
| 1218 |
+
"eval_steps_per_second": 0.612,
|
| 1219 |
+
"step": 5500
|
| 1220 |
+
},
|
| 1221 |
+
{
|
| 1222 |
+
"epoch": 2.825865580448065,
|
| 1223 |
+
"grad_norm": 1.214141845703125,
|
| 1224 |
+
"learning_rate": 1e-05,
|
| 1225 |
+
"loss": 0.2406,
|
| 1226 |
+
"step": 5550
|
| 1227 |
+
},
|
| 1228 |
+
{
|
| 1229 |
+
"epoch": 2.8513238289205702,
|
| 1230 |
+
"grad_norm": 1.3861531019210815,
|
| 1231 |
+
"learning_rate": 1e-05,
|
| 1232 |
+
"loss": 0.2479,
|
| 1233 |
+
"step": 5600
|
| 1234 |
+
},
|
| 1235 |
+
{
|
| 1236 |
+
"epoch": 2.8513238289205702,
|
| 1237 |
+
"eval_loss": 0.24935197830200195,
|
| 1238 |
+
"eval_runtime": 21.4872,
|
| 1239 |
+
"eval_samples_per_second": 4.654,
|
| 1240 |
+
"eval_steps_per_second": 0.605,
|
| 1241 |
+
"step": 5600
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 2.8767820773930755,
|
| 1245 |
+
"grad_norm": 1.1861828565597534,
|
| 1246 |
+
"learning_rate": 1e-05,
|
| 1247 |
+
"loss": 0.2318,
|
| 1248 |
+
"step": 5650
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 2.9022403258655807,
|
| 1252 |
+
"grad_norm": 1.1281813383102417,
|
| 1253 |
+
"learning_rate": 1e-05,
|
| 1254 |
+
"loss": 0.2413,
|
| 1255 |
+
"step": 5700
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 2.9022403258655807,
|
| 1259 |
+
"eval_loss": 0.25039103627204895,
|
| 1260 |
+
"eval_runtime": 20.9675,
|
| 1261 |
+
"eval_samples_per_second": 4.769,
|
| 1262 |
+
"eval_steps_per_second": 0.62,
|
| 1263 |
+
"step": 5700
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 2.9276985743380854,
|
| 1267 |
+
"grad_norm": 1.1463491916656494,
|
| 1268 |
+
"learning_rate": 1e-05,
|
| 1269 |
+
"loss": 0.2364,
|
| 1270 |
+
"step": 5750
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 2.9531568228105907,
|
| 1274 |
+
"grad_norm": 1.3118984699249268,
|
| 1275 |
+
"learning_rate": 1e-05,
|
| 1276 |
+
"loss": 0.2388,
|
| 1277 |
+
"step": 5800
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 2.9531568228105907,
|
| 1281 |
+
"eval_loss": 0.24982962012290955,
|
| 1282 |
+
"eval_runtime": 21.5908,
|
| 1283 |
+
"eval_samples_per_second": 4.632,
|
| 1284 |
+
"eval_steps_per_second": 0.602,
|
| 1285 |
+
"step": 5800
|
| 1286 |
+
},
|
| 1287 |
+
{
|
| 1288 |
+
"epoch": 2.978615071283096,
|
| 1289 |
+
"grad_norm": 1.0851026773452759,
|
| 1290 |
+
"learning_rate": 1e-05,
|
| 1291 |
+
"loss": 0.2348,
|
| 1292 |
+
"step": 5850
|
| 1293 |
+
}
|
| 1294 |
+
],
|
| 1295 |
+
"logging_steps": 50,
|
| 1296 |
+
"max_steps": 5891,
|
| 1297 |
+
"num_input_tokens_seen": 0,
|
| 1298 |
+
"num_train_epochs": 3,
|
| 1299 |
+
"save_steps": 100,
|
| 1300 |
+
"stateful_callbacks": {
|
| 1301 |
+
"TrainerControl": {
|
| 1302 |
+
"args": {
|
| 1303 |
+
"should_epoch_stop": false,
|
| 1304 |
+
"should_evaluate": false,
|
| 1305 |
+
"should_log": false,
|
| 1306 |
+
"should_save": true,
|
| 1307 |
+
"should_training_stop": true
|
| 1308 |
+
},
|
| 1309 |
+
"attributes": {}
|
| 1310 |
+
}
|
| 1311 |
+
},
|
| 1312 |
+
"total_flos": 6.421089311391744e+20,
|
| 1313 |
+
"train_batch_size": 16,
|
| 1314 |
+
"trial_name": null,
|
| 1315 |
+
"trial_params": null
|
| 1316 |
+
}
|