Upload folder using huggingface_hub
Browse files- optimizer.pt +3 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- trainer_state.json +2512 -0
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:094959f1f69c2c2ef8a488c51689aedce1875ecddadda8fc0b2165a600b8ebc1
|
| 3 |
+
size 531043898
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d2256391bb0522487c9086ceda84962369530511a9335436aa9df2ec8ea23fca
|
| 3 |
+
size 14244
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:dd2aba385088a560af280b575da41b2caeea370ace48f1d5e2b7ce2baba0491c
|
| 3 |
+
size 1064
|
trainer_state.json
ADDED
|
@@ -0,0 +1,2512 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 3.0,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 3546,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.008460236886632826,
|
| 14 |
+
"grad_norm": 2.4233109951019287,
|
| 15 |
+
"learning_rate": 1.9949238578680207e-05,
|
| 16 |
+
"loss": 1.3216,
|
| 17 |
+
"step": 10
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.01692047377326565,
|
| 21 |
+
"grad_norm": 1.1721662282943726,
|
| 22 |
+
"learning_rate": 1.9892836999435986e-05,
|
| 23 |
+
"loss": 0.5312,
|
| 24 |
+
"step": 20
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.025380710659898477,
|
| 28 |
+
"grad_norm": 0.899889349937439,
|
| 29 |
+
"learning_rate": 1.9836435420191765e-05,
|
| 30 |
+
"loss": 0.2969,
|
| 31 |
+
"step": 30
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.0338409475465313,
|
| 35 |
+
"grad_norm": 0.4674186408519745,
|
| 36 |
+
"learning_rate": 1.9780033840947548e-05,
|
| 37 |
+
"loss": 0.1564,
|
| 38 |
+
"step": 40
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.04230118443316413,
|
| 42 |
+
"grad_norm": 0.9013752937316895,
|
| 43 |
+
"learning_rate": 1.972363226170333e-05,
|
| 44 |
+
"loss": 0.0966,
|
| 45 |
+
"step": 50
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.050761421319796954,
|
| 49 |
+
"grad_norm": 0.364164799451828,
|
| 50 |
+
"learning_rate": 1.966723068245911e-05,
|
| 51 |
+
"loss": 0.0516,
|
| 52 |
+
"step": 60
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.05922165820642978,
|
| 56 |
+
"grad_norm": 0.22717136144638062,
|
| 57 |
+
"learning_rate": 1.9610829103214893e-05,
|
| 58 |
+
"loss": 0.0307,
|
| 59 |
+
"step": 70
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.0676818950930626,
|
| 63 |
+
"grad_norm": 0.09149003773927689,
|
| 64 |
+
"learning_rate": 1.9554427523970672e-05,
|
| 65 |
+
"loss": 0.0182,
|
| 66 |
+
"step": 80
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.07614213197969544,
|
| 70 |
+
"grad_norm": 0.12942063808441162,
|
| 71 |
+
"learning_rate": 1.949802594472645e-05,
|
| 72 |
+
"loss": 0.017,
|
| 73 |
+
"step": 90
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.08460236886632826,
|
| 77 |
+
"grad_norm": 0.27440059185028076,
|
| 78 |
+
"learning_rate": 1.9441624365482234e-05,
|
| 79 |
+
"loss": 0.0113,
|
| 80 |
+
"step": 100
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.09306260575296109,
|
| 84 |
+
"grad_norm": 0.5005059242248535,
|
| 85 |
+
"learning_rate": 1.9385222786238017e-05,
|
| 86 |
+
"loss": 0.0082,
|
| 87 |
+
"step": 110
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.10152284263959391,
|
| 91 |
+
"grad_norm": 0.19820384681224823,
|
| 92 |
+
"learning_rate": 1.9328821206993796e-05,
|
| 93 |
+
"loss": 0.0057,
|
| 94 |
+
"step": 120
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.10998307952622674,
|
| 98 |
+
"grad_norm": 0.03431914001703262,
|
| 99 |
+
"learning_rate": 1.927241962774958e-05,
|
| 100 |
+
"loss": 0.0055,
|
| 101 |
+
"step": 130
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.11844331641285956,
|
| 105 |
+
"grad_norm": 0.4414117634296417,
|
| 106 |
+
"learning_rate": 1.921601804850536e-05,
|
| 107 |
+
"loss": 0.0063,
|
| 108 |
+
"step": 140
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.12690355329949238,
|
| 112 |
+
"grad_norm": 0.9055352807044983,
|
| 113 |
+
"learning_rate": 1.915961646926114e-05,
|
| 114 |
+
"loss": 0.0048,
|
| 115 |
+
"step": 150
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.1353637901861252,
|
| 119 |
+
"grad_norm": 0.036724768579006195,
|
| 120 |
+
"learning_rate": 1.910321489001692e-05,
|
| 121 |
+
"loss": 0.0035,
|
| 122 |
+
"step": 160
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.14382402707275804,
|
| 126 |
+
"grad_norm": 0.0888664722442627,
|
| 127 |
+
"learning_rate": 1.9046813310772703e-05,
|
| 128 |
+
"loss": 0.0033,
|
| 129 |
+
"step": 170
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.15228426395939088,
|
| 133 |
+
"grad_norm": 0.019098607823252678,
|
| 134 |
+
"learning_rate": 1.8990411731528486e-05,
|
| 135 |
+
"loss": 0.0029,
|
| 136 |
+
"step": 180
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.16074450084602368,
|
| 140 |
+
"grad_norm": 0.0171552412211895,
|
| 141 |
+
"learning_rate": 1.8934010152284265e-05,
|
| 142 |
+
"loss": 0.0029,
|
| 143 |
+
"step": 190
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.1692047377326565,
|
| 147 |
+
"grad_norm": 0.01885647512972355,
|
| 148 |
+
"learning_rate": 1.8877608573040048e-05,
|
| 149 |
+
"loss": 0.0028,
|
| 150 |
+
"step": 200
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.17766497461928935,
|
| 154 |
+
"grad_norm": 0.01599389687180519,
|
| 155 |
+
"learning_rate": 1.8821206993795827e-05,
|
| 156 |
+
"loss": 0.0022,
|
| 157 |
+
"step": 210
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.18612521150592218,
|
| 161 |
+
"grad_norm": 0.017242038622498512,
|
| 162 |
+
"learning_rate": 1.876480541455161e-05,
|
| 163 |
+
"loss": 0.0023,
|
| 164 |
+
"step": 220
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.19458544839255498,
|
| 168 |
+
"grad_norm": 0.5941590666770935,
|
| 169 |
+
"learning_rate": 1.870840383530739e-05,
|
| 170 |
+
"loss": 0.0031,
|
| 171 |
+
"step": 230
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.20304568527918782,
|
| 175 |
+
"grad_norm": 0.4872148931026459,
|
| 176 |
+
"learning_rate": 1.8652002256063172e-05,
|
| 177 |
+
"loss": 0.0027,
|
| 178 |
+
"step": 240
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.21150592216582065,
|
| 182 |
+
"grad_norm": 0.014173777773976326,
|
| 183 |
+
"learning_rate": 1.859560067681895e-05,
|
| 184 |
+
"loss": 0.0021,
|
| 185 |
+
"step": 250
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.21996615905245348,
|
| 189 |
+
"grad_norm": 0.024433018639683723,
|
| 190 |
+
"learning_rate": 1.8539199097574734e-05,
|
| 191 |
+
"loss": 0.0017,
|
| 192 |
+
"step": 260
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.22842639593908629,
|
| 196 |
+
"grad_norm": 0.017857005819678307,
|
| 197 |
+
"learning_rate": 1.8482797518330516e-05,
|
| 198 |
+
"loss": 0.0017,
|
| 199 |
+
"step": 270
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.23688663282571912,
|
| 203 |
+
"grad_norm": 0.014704135246574879,
|
| 204 |
+
"learning_rate": 1.8426395939086296e-05,
|
| 205 |
+
"loss": 0.0021,
|
| 206 |
+
"step": 280
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.24534686971235195,
|
| 210 |
+
"grad_norm": 0.022361995652318,
|
| 211 |
+
"learning_rate": 1.8369994359842075e-05,
|
| 212 |
+
"loss": 0.0016,
|
| 213 |
+
"step": 290
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.25380710659898476,
|
| 217 |
+
"grad_norm": 0.01112055778503418,
|
| 218 |
+
"learning_rate": 1.8313592780597858e-05,
|
| 219 |
+
"loss": 0.0015,
|
| 220 |
+
"step": 300
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.2622673434856176,
|
| 224 |
+
"grad_norm": 0.010048450902104378,
|
| 225 |
+
"learning_rate": 1.825719120135364e-05,
|
| 226 |
+
"loss": 0.0015,
|
| 227 |
+
"step": 310
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.2707275803722504,
|
| 231 |
+
"grad_norm": 0.012935510836541653,
|
| 232 |
+
"learning_rate": 1.820078962210942e-05,
|
| 233 |
+
"loss": 0.0014,
|
| 234 |
+
"step": 320
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.27918781725888325,
|
| 238 |
+
"grad_norm": 0.010541570372879505,
|
| 239 |
+
"learning_rate": 1.8144388042865203e-05,
|
| 240 |
+
"loss": 0.0013,
|
| 241 |
+
"step": 330
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.2876480541455161,
|
| 245 |
+
"grad_norm": 0.009258048608899117,
|
| 246 |
+
"learning_rate": 1.8087986463620982e-05,
|
| 247 |
+
"loss": 0.0012,
|
| 248 |
+
"step": 340
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.2961082910321489,
|
| 252 |
+
"grad_norm": 0.009454768151044846,
|
| 253 |
+
"learning_rate": 1.8031584884376765e-05,
|
| 254 |
+
"loss": 0.0015,
|
| 255 |
+
"step": 350
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.30456852791878175,
|
| 259 |
+
"grad_norm": 0.018366724252700806,
|
| 260 |
+
"learning_rate": 1.7975183305132544e-05,
|
| 261 |
+
"loss": 0.0012,
|
| 262 |
+
"step": 360
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.3130287648054145,
|
| 266 |
+
"grad_norm": 0.009252658113837242,
|
| 267 |
+
"learning_rate": 1.7918781725888327e-05,
|
| 268 |
+
"loss": 0.0011,
|
| 269 |
+
"step": 370
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.32148900169204736,
|
| 273 |
+
"grad_norm": 0.1430797576904297,
|
| 274 |
+
"learning_rate": 1.7862380146644106e-05,
|
| 275 |
+
"loss": 0.0018,
|
| 276 |
+
"step": 380
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.3299492385786802,
|
| 280 |
+
"grad_norm": 0.009182159788906574,
|
| 281 |
+
"learning_rate": 1.780597856739989e-05,
|
| 282 |
+
"loss": 0.0018,
|
| 283 |
+
"step": 390
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.338409475465313,
|
| 287 |
+
"grad_norm": 0.00843009352684021,
|
| 288 |
+
"learning_rate": 1.774957698815567e-05,
|
| 289 |
+
"loss": 0.0017,
|
| 290 |
+
"step": 400
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.34686971235194586,
|
| 294 |
+
"grad_norm": 0.008176930248737335,
|
| 295 |
+
"learning_rate": 1.769317540891145e-05,
|
| 296 |
+
"loss": 0.0013,
|
| 297 |
+
"step": 410
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.3553299492385787,
|
| 301 |
+
"grad_norm": 0.008539380505681038,
|
| 302 |
+
"learning_rate": 1.7636773829667233e-05,
|
| 303 |
+
"loss": 0.001,
|
| 304 |
+
"step": 420
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.3637901861252115,
|
| 308 |
+
"grad_norm": 0.008693977259099483,
|
| 309 |
+
"learning_rate": 1.7580372250423013e-05,
|
| 310 |
+
"loss": 0.0014,
|
| 311 |
+
"step": 430
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.37225042301184436,
|
| 315 |
+
"grad_norm": 0.009141940623521805,
|
| 316 |
+
"learning_rate": 1.7523970671178795e-05,
|
| 317 |
+
"loss": 0.0012,
|
| 318 |
+
"step": 440
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.38071065989847713,
|
| 322 |
+
"grad_norm": 0.009133282117545605,
|
| 323 |
+
"learning_rate": 1.7467569091934575e-05,
|
| 324 |
+
"loss": 0.0015,
|
| 325 |
+
"step": 450
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.38917089678510997,
|
| 329 |
+
"grad_norm": 0.5799979567527771,
|
| 330 |
+
"learning_rate": 1.7411167512690357e-05,
|
| 331 |
+
"loss": 0.0013,
|
| 332 |
+
"step": 460
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.3976311336717428,
|
| 336 |
+
"grad_norm": 0.04120843857526779,
|
| 337 |
+
"learning_rate": 1.7354765933446137e-05,
|
| 338 |
+
"loss": 0.0011,
|
| 339 |
+
"step": 470
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.40609137055837563,
|
| 343 |
+
"grad_norm": 0.01276948768645525,
|
| 344 |
+
"learning_rate": 1.729836435420192e-05,
|
| 345 |
+
"loss": 0.0012,
|
| 346 |
+
"step": 480
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.41455160744500846,
|
| 350 |
+
"grad_norm": 0.0076572224497795105,
|
| 351 |
+
"learning_rate": 1.7241962774957702e-05,
|
| 352 |
+
"loss": 0.0014,
|
| 353 |
+
"step": 490
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.4230118443316413,
|
| 357 |
+
"grad_norm": 0.008293437771499157,
|
| 358 |
+
"learning_rate": 1.718556119571348e-05,
|
| 359 |
+
"loss": 0.0009,
|
| 360 |
+
"step": 500
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.43147208121827413,
|
| 364 |
+
"grad_norm": 0.0068985214456915855,
|
| 365 |
+
"learning_rate": 1.712915961646926e-05,
|
| 366 |
+
"loss": 0.0009,
|
| 367 |
+
"step": 510
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.43993231810490696,
|
| 371 |
+
"grad_norm": 0.029666345566511154,
|
| 372 |
+
"learning_rate": 1.7072758037225043e-05,
|
| 373 |
+
"loss": 0.0009,
|
| 374 |
+
"step": 520
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.44839255499153974,
|
| 378 |
+
"grad_norm": 0.014065027236938477,
|
| 379 |
+
"learning_rate": 1.7016356457980826e-05,
|
| 380 |
+
"loss": 0.0008,
|
| 381 |
+
"step": 530
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.45685279187817257,
|
| 385 |
+
"grad_norm": 0.00650499714538455,
|
| 386 |
+
"learning_rate": 1.6959954878736606e-05,
|
| 387 |
+
"loss": 0.0008,
|
| 388 |
+
"step": 540
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.4653130287648054,
|
| 392 |
+
"grad_norm": 0.0065047661773860455,
|
| 393 |
+
"learning_rate": 1.6903553299492388e-05,
|
| 394 |
+
"loss": 0.0008,
|
| 395 |
+
"step": 550
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.47377326565143824,
|
| 399 |
+
"grad_norm": 0.0071571446023881435,
|
| 400 |
+
"learning_rate": 1.684715172024817e-05,
|
| 401 |
+
"loss": 0.0008,
|
| 402 |
+
"step": 560
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.48223350253807107,
|
| 406 |
+
"grad_norm": 0.005875805858522654,
|
| 407 |
+
"learning_rate": 1.679075014100395e-05,
|
| 408 |
+
"loss": 0.0008,
|
| 409 |
+
"step": 570
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.4906937394247039,
|
| 413 |
+
"grad_norm": 0.005848041269928217,
|
| 414 |
+
"learning_rate": 1.673434856175973e-05,
|
| 415 |
+
"loss": 0.0007,
|
| 416 |
+
"step": 580
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.49915397631133673,
|
| 420 |
+
"grad_norm": 0.006032396107912064,
|
| 421 |
+
"learning_rate": 1.6677946982515512e-05,
|
| 422 |
+
"loss": 0.0007,
|
| 423 |
+
"step": 590
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.5076142131979695,
|
| 427 |
+
"grad_norm": 0.005858593620359898,
|
| 428 |
+
"learning_rate": 1.662154540327129e-05,
|
| 429 |
+
"loss": 0.0007,
|
| 430 |
+
"step": 600
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.5160744500846024,
|
| 434 |
+
"grad_norm": 0.006055026315152645,
|
| 435 |
+
"learning_rate": 1.6565143824027074e-05,
|
| 436 |
+
"loss": 0.0007,
|
| 437 |
+
"step": 610
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.5245346869712352,
|
| 441 |
+
"grad_norm": 0.0055991546250879765,
|
| 442 |
+
"learning_rate": 1.6508742244782857e-05,
|
| 443 |
+
"loss": 0.0039,
|
| 444 |
+
"step": 620
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.5329949238578681,
|
| 448 |
+
"grad_norm": 0.006043154280632734,
|
| 449 |
+
"learning_rate": 1.6452340665538636e-05,
|
| 450 |
+
"loss": 0.0009,
|
| 451 |
+
"step": 630
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.5414551607445008,
|
| 455 |
+
"grad_norm": 0.006406570319086313,
|
| 456 |
+
"learning_rate": 1.6395939086294416e-05,
|
| 457 |
+
"loss": 0.0008,
|
| 458 |
+
"step": 640
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.5499153976311336,
|
| 462 |
+
"grad_norm": 0.006317532621324062,
|
| 463 |
+
"learning_rate": 1.63395375070502e-05,
|
| 464 |
+
"loss": 0.0007,
|
| 465 |
+
"step": 650
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.5583756345177665,
|
| 469 |
+
"grad_norm": 0.0260737594217062,
|
| 470 |
+
"learning_rate": 1.628313592780598e-05,
|
| 471 |
+
"loss": 0.0007,
|
| 472 |
+
"step": 660
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.5668358714043993,
|
| 476 |
+
"grad_norm": 0.005516720470041037,
|
| 477 |
+
"learning_rate": 1.622673434856176e-05,
|
| 478 |
+
"loss": 0.0007,
|
| 479 |
+
"step": 670
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 0.5752961082910322,
|
| 483 |
+
"grad_norm": 0.005611425265669823,
|
| 484 |
+
"learning_rate": 1.6170332769317543e-05,
|
| 485 |
+
"loss": 0.0006,
|
| 486 |
+
"step": 680
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.583756345177665,
|
| 490 |
+
"grad_norm": 0.005698953289538622,
|
| 491 |
+
"learning_rate": 1.6113931190073326e-05,
|
| 492 |
+
"loss": 0.0006,
|
| 493 |
+
"step": 690
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.5922165820642978,
|
| 497 |
+
"grad_norm": 0.005045793950557709,
|
| 498 |
+
"learning_rate": 1.6057529610829105e-05,
|
| 499 |
+
"loss": 0.0006,
|
| 500 |
+
"step": 700
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.6006768189509306,
|
| 504 |
+
"grad_norm": 0.005404220428317785,
|
| 505 |
+
"learning_rate": 1.6001128031584884e-05,
|
| 506 |
+
"loss": 0.0007,
|
| 507 |
+
"step": 710
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 0.6091370558375635,
|
| 511 |
+
"grad_norm": 0.005528539884835482,
|
| 512 |
+
"learning_rate": 1.5944726452340667e-05,
|
| 513 |
+
"loss": 0.0007,
|
| 514 |
+
"step": 720
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.6175972927241963,
|
| 518 |
+
"grad_norm": 0.0052613625302910805,
|
| 519 |
+
"learning_rate": 1.5888324873096446e-05,
|
| 520 |
+
"loss": 0.0006,
|
| 521 |
+
"step": 730
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.626057529610829,
|
| 525 |
+
"grad_norm": 0.0050328257493674755,
|
| 526 |
+
"learning_rate": 1.583192329385223e-05,
|
| 527 |
+
"loss": 0.0006,
|
| 528 |
+
"step": 740
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.6345177664974619,
|
| 532 |
+
"grad_norm": 0.005106599070131779,
|
| 533 |
+
"learning_rate": 1.5775521714608012e-05,
|
| 534 |
+
"loss": 0.0006,
|
| 535 |
+
"step": 750
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.6429780033840947,
|
| 539 |
+
"grad_norm": 0.004836504813283682,
|
| 540 |
+
"learning_rate": 1.571912013536379e-05,
|
| 541 |
+
"loss": 0.0006,
|
| 542 |
+
"step": 760
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.6514382402707276,
|
| 546 |
+
"grad_norm": 0.004986999090760946,
|
| 547 |
+
"learning_rate": 1.566271855611957e-05,
|
| 548 |
+
"loss": 0.0006,
|
| 549 |
+
"step": 770
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.6598984771573604,
|
| 553 |
+
"grad_norm": 0.005235906690359116,
|
| 554 |
+
"learning_rate": 1.5606316976875353e-05,
|
| 555 |
+
"loss": 0.0005,
|
| 556 |
+
"step": 780
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.6683587140439933,
|
| 560 |
+
"grad_norm": 0.4625732898712158,
|
| 561 |
+
"learning_rate": 1.5549915397631136e-05,
|
| 562 |
+
"loss": 0.0009,
|
| 563 |
+
"step": 790
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.676818950930626,
|
| 567 |
+
"grad_norm": 0.004936546087265015,
|
| 568 |
+
"learning_rate": 1.5493513818386915e-05,
|
| 569 |
+
"loss": 0.0006,
|
| 570 |
+
"step": 800
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.6852791878172588,
|
| 574 |
+
"grad_norm": 0.004956043791025877,
|
| 575 |
+
"learning_rate": 1.5437112239142698e-05,
|
| 576 |
+
"loss": 0.0005,
|
| 577 |
+
"step": 810
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.6937394247038917,
|
| 581 |
+
"grad_norm": 0.005084918346256018,
|
| 582 |
+
"learning_rate": 1.538071065989848e-05,
|
| 583 |
+
"loss": 0.0005,
|
| 584 |
+
"step": 820
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.7021996615905245,
|
| 588 |
+
"grad_norm": 0.006050592288374901,
|
| 589 |
+
"learning_rate": 1.532430908065426e-05,
|
| 590 |
+
"loss": 0.0005,
|
| 591 |
+
"step": 830
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 0.7106598984771574,
|
| 595 |
+
"grad_norm": 0.0045923274010419846,
|
| 596 |
+
"learning_rate": 1.526790750141004e-05,
|
| 597 |
+
"loss": 0.0005,
|
| 598 |
+
"step": 840
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 0.7191201353637902,
|
| 602 |
+
"grad_norm": 0.004406214691698551,
|
| 603 |
+
"learning_rate": 1.5211505922165822e-05,
|
| 604 |
+
"loss": 0.0005,
|
| 605 |
+
"step": 850
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 0.727580372250423,
|
| 609 |
+
"grad_norm": 0.00449965288862586,
|
| 610 |
+
"learning_rate": 1.5155104342921603e-05,
|
| 611 |
+
"loss": 0.0005,
|
| 612 |
+
"step": 860
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 0.7360406091370558,
|
| 616 |
+
"grad_norm": 0.00427517294883728,
|
| 617 |
+
"learning_rate": 1.5098702763677384e-05,
|
| 618 |
+
"loss": 0.0006,
|
| 619 |
+
"step": 870
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 0.7445008460236887,
|
| 623 |
+
"grad_norm": 0.006013456266373396,
|
| 624 |
+
"learning_rate": 1.5042301184433165e-05,
|
| 625 |
+
"loss": 0.0005,
|
| 626 |
+
"step": 880
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.7529610829103215,
|
| 630 |
+
"grad_norm": 0.004456107504665852,
|
| 631 |
+
"learning_rate": 1.4985899605188948e-05,
|
| 632 |
+
"loss": 0.0005,
|
| 633 |
+
"step": 890
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.7614213197969543,
|
| 637 |
+
"grad_norm": 0.004223175812512636,
|
| 638 |
+
"learning_rate": 1.4929498025944729e-05,
|
| 639 |
+
"loss": 0.0005,
|
| 640 |
+
"step": 900
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.7698815566835872,
|
| 644 |
+
"grad_norm": 0.004154487047344446,
|
| 645 |
+
"learning_rate": 1.4873096446700508e-05,
|
| 646 |
+
"loss": 0.0005,
|
| 647 |
+
"step": 910
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.7783417935702199,
|
| 651 |
+
"grad_norm": 0.007055574096739292,
|
| 652 |
+
"learning_rate": 1.4816694867456289e-05,
|
| 653 |
+
"loss": 0.0005,
|
| 654 |
+
"step": 920
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 0.7868020304568528,
|
| 658 |
+
"grad_norm": 0.005269153974950314,
|
| 659 |
+
"learning_rate": 1.476029328821207e-05,
|
| 660 |
+
"loss": 0.0005,
|
| 661 |
+
"step": 930
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 0.7952622673434856,
|
| 665 |
+
"grad_norm": 0.004041856154799461,
|
| 666 |
+
"learning_rate": 1.4703891708967853e-05,
|
| 667 |
+
"loss": 0.0005,
|
| 668 |
+
"step": 940
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.8037225042301185,
|
| 672 |
+
"grad_norm": 0.003965588286519051,
|
| 673 |
+
"learning_rate": 1.4647490129723634e-05,
|
| 674 |
+
"loss": 0.0005,
|
| 675 |
+
"step": 950
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 0.8121827411167513,
|
| 679 |
+
"grad_norm": 0.005661304574459791,
|
| 680 |
+
"learning_rate": 1.4591088550479415e-05,
|
| 681 |
+
"loss": 0.0005,
|
| 682 |
+
"step": 960
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 0.8206429780033841,
|
| 686 |
+
"grad_norm": 0.0039055896922945976,
|
| 687 |
+
"learning_rate": 1.4534686971235198e-05,
|
| 688 |
+
"loss": 0.0011,
|
| 689 |
+
"step": 970
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.8291032148900169,
|
| 693 |
+
"grad_norm": 1.8310741186141968,
|
| 694 |
+
"learning_rate": 1.4478285391990977e-05,
|
| 695 |
+
"loss": 0.001,
|
| 696 |
+
"step": 980
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 0.8375634517766497,
|
| 700 |
+
"grad_norm": 0.005107107572257519,
|
| 701 |
+
"learning_rate": 1.4421883812746758e-05,
|
| 702 |
+
"loss": 0.0008,
|
| 703 |
+
"step": 990
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 0.8460236886632826,
|
| 707 |
+
"grad_norm": 0.005387498531490564,
|
| 708 |
+
"learning_rate": 1.4365482233502539e-05,
|
| 709 |
+
"loss": 0.0005,
|
| 710 |
+
"step": 1000
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 0.8544839255499154,
|
| 714 |
+
"grad_norm": 0.0051966467872262,
|
| 715 |
+
"learning_rate": 1.430908065425832e-05,
|
| 716 |
+
"loss": 0.0005,
|
| 717 |
+
"step": 1010
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 0.8629441624365483,
|
| 721 |
+
"grad_norm": 0.004603747744113207,
|
| 722 |
+
"learning_rate": 1.4252679075014103e-05,
|
| 723 |
+
"loss": 0.0005,
|
| 724 |
+
"step": 1020
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 0.871404399323181,
|
| 728 |
+
"grad_norm": 0.0113060986623168,
|
| 729 |
+
"learning_rate": 1.4196277495769884e-05,
|
| 730 |
+
"loss": 0.0005,
|
| 731 |
+
"step": 1030
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 0.8798646362098139,
|
| 735 |
+
"grad_norm": 0.004159980919212103,
|
| 736 |
+
"learning_rate": 1.4139875916525665e-05,
|
| 737 |
+
"loss": 0.0004,
|
| 738 |
+
"step": 1040
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 0.8883248730964467,
|
| 742 |
+
"grad_norm": 0.004094019532203674,
|
| 743 |
+
"learning_rate": 1.4083474337281444e-05,
|
| 744 |
+
"loss": 0.0004,
|
| 745 |
+
"step": 1050
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 0.8967851099830795,
|
| 749 |
+
"grad_norm": 0.00435257563367486,
|
| 750 |
+
"learning_rate": 1.4027072758037225e-05,
|
| 751 |
+
"loss": 0.0004,
|
| 752 |
+
"step": 1060
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 0.9052453468697124,
|
| 756 |
+
"grad_norm": 0.003843541955575347,
|
| 757 |
+
"learning_rate": 1.3970671178793008e-05,
|
| 758 |
+
"loss": 0.0004,
|
| 759 |
+
"step": 1070
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 0.9137055837563451,
|
| 763 |
+
"grad_norm": 0.16885782778263092,
|
| 764 |
+
"learning_rate": 1.3914269599548789e-05,
|
| 765 |
+
"loss": 0.0005,
|
| 766 |
+
"step": 1080
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 0.922165820642978,
|
| 770 |
+
"grad_norm": 0.0037802334409207106,
|
| 771 |
+
"learning_rate": 1.385786802030457e-05,
|
| 772 |
+
"loss": 0.0004,
|
| 773 |
+
"step": 1090
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 0.9306260575296108,
|
| 777 |
+
"grad_norm": 0.0037738329265266657,
|
| 778 |
+
"learning_rate": 1.3801466441060352e-05,
|
| 779 |
+
"loss": 0.0005,
|
| 780 |
+
"step": 1100
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 0.9390862944162437,
|
| 784 |
+
"grad_norm": 0.012493060901761055,
|
| 785 |
+
"learning_rate": 1.374506486181613e-05,
|
| 786 |
+
"loss": 0.0004,
|
| 787 |
+
"step": 1110
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 0.9475465313028765,
|
| 791 |
+
"grad_norm": 0.0037303464487195015,
|
| 792 |
+
"learning_rate": 1.3688663282571913e-05,
|
| 793 |
+
"loss": 0.0004,
|
| 794 |
+
"step": 1120
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 0.9560067681895094,
|
| 798 |
+
"grad_norm": 0.0035704418551176786,
|
| 799 |
+
"learning_rate": 1.3632261703327694e-05,
|
| 800 |
+
"loss": 0.0007,
|
| 801 |
+
"step": 1130
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 0.9644670050761421,
|
| 805 |
+
"grad_norm": 0.0038276014383882284,
|
| 806 |
+
"learning_rate": 1.3575860124083475e-05,
|
| 807 |
+
"loss": 0.0004,
|
| 808 |
+
"step": 1140
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 0.9729272419627749,
|
| 812 |
+
"grad_norm": 0.004172508604824543,
|
| 813 |
+
"learning_rate": 1.3519458544839257e-05,
|
| 814 |
+
"loss": 0.0004,
|
| 815 |
+
"step": 1150
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 0.9813874788494078,
|
| 819 |
+
"grad_norm": 0.20282192528247833,
|
| 820 |
+
"learning_rate": 1.3463056965595038e-05,
|
| 821 |
+
"loss": 0.0004,
|
| 822 |
+
"step": 1160
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 0.9898477157360406,
|
| 826 |
+
"grad_norm": 0.0035703997127711773,
|
| 827 |
+
"learning_rate": 1.340665538635082e-05,
|
| 828 |
+
"loss": 0.0004,
|
| 829 |
+
"step": 1170
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 0.9983079526226735,
|
| 833 |
+
"grad_norm": 0.003505186177790165,
|
| 834 |
+
"learning_rate": 1.3350253807106599e-05,
|
| 835 |
+
"loss": 0.0004,
|
| 836 |
+
"step": 1180
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 1.0067681895093064,
|
| 840 |
+
"grad_norm": 0.003632462117820978,
|
| 841 |
+
"learning_rate": 1.329385222786238e-05,
|
| 842 |
+
"loss": 0.0004,
|
| 843 |
+
"step": 1190
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 1.015228426395939,
|
| 847 |
+
"grad_norm": 0.0034904363565146923,
|
| 848 |
+
"learning_rate": 1.3237450648618163e-05,
|
| 849 |
+
"loss": 0.0004,
|
| 850 |
+
"step": 1200
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 1.023688663282572,
|
| 854 |
+
"grad_norm": 0.003400213085114956,
|
| 855 |
+
"learning_rate": 1.3181049069373944e-05,
|
| 856 |
+
"loss": 0.0004,
|
| 857 |
+
"step": 1210
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 1.0321489001692048,
|
| 861 |
+
"grad_norm": 0.02098667249083519,
|
| 862 |
+
"learning_rate": 1.3124647490129725e-05,
|
| 863 |
+
"loss": 0.0004,
|
| 864 |
+
"step": 1220
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 1.0406091370558375,
|
| 868 |
+
"grad_norm": 0.0033113211393356323,
|
| 869 |
+
"learning_rate": 1.3068245910885506e-05,
|
| 870 |
+
"loss": 0.0004,
|
| 871 |
+
"step": 1230
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 1.0490693739424704,
|
| 875 |
+
"grad_norm": 0.0075438846834003925,
|
| 876 |
+
"learning_rate": 1.3011844331641288e-05,
|
| 877 |
+
"loss": 0.001,
|
| 878 |
+
"step": 1240
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 1.0575296108291032,
|
| 882 |
+
"grad_norm": 0.003400578862056136,
|
| 883 |
+
"learning_rate": 1.2955442752397068e-05,
|
| 884 |
+
"loss": 0.0006,
|
| 885 |
+
"step": 1250
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 1.0659898477157361,
|
| 889 |
+
"grad_norm": 0.003571244655176997,
|
| 890 |
+
"learning_rate": 1.2899041173152849e-05,
|
| 891 |
+
"loss": 0.0004,
|
| 892 |
+
"step": 1260
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 1.0744500846023688,
|
| 896 |
+
"grad_norm": 0.0032794324215501547,
|
| 897 |
+
"learning_rate": 1.284263959390863e-05,
|
| 898 |
+
"loss": 0.0004,
|
| 899 |
+
"step": 1270
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 1.0829103214890017,
|
| 903 |
+
"grad_norm": 0.0032312178518623114,
|
| 904 |
+
"learning_rate": 1.2786238014664412e-05,
|
| 905 |
+
"loss": 0.0011,
|
| 906 |
+
"step": 1280
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 1.0913705583756346,
|
| 910 |
+
"grad_norm": 0.003224333981052041,
|
| 911 |
+
"learning_rate": 1.2729836435420193e-05,
|
| 912 |
+
"loss": 0.0003,
|
| 913 |
+
"step": 1290
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 1.0998307952622675,
|
| 917 |
+
"grad_norm": 0.0033167279325425625,
|
| 918 |
+
"learning_rate": 1.2673434856175974e-05,
|
| 919 |
+
"loss": 0.0004,
|
| 920 |
+
"step": 1300
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 1.1082910321489001,
|
| 924 |
+
"grad_norm": 0.003647689940407872,
|
| 925 |
+
"learning_rate": 1.2617033276931755e-05,
|
| 926 |
+
"loss": 0.0009,
|
| 927 |
+
"step": 1310
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 1.116751269035533,
|
| 931 |
+
"grad_norm": 0.005791259463876486,
|
| 932 |
+
"learning_rate": 1.2560631697687535e-05,
|
| 933 |
+
"loss": 0.0004,
|
| 934 |
+
"step": 1320
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 1.125211505922166,
|
| 938 |
+
"grad_norm": 0.0036105727776885033,
|
| 939 |
+
"learning_rate": 1.2504230118443317e-05,
|
| 940 |
+
"loss": 0.0005,
|
| 941 |
+
"step": 1330
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 1.1336717428087986,
|
| 945 |
+
"grad_norm": 0.0033262548968195915,
|
| 946 |
+
"learning_rate": 1.2447828539199098e-05,
|
| 947 |
+
"loss": 0.0003,
|
| 948 |
+
"step": 1340
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 1.1421319796954315,
|
| 952 |
+
"grad_norm": 0.003255200106650591,
|
| 953 |
+
"learning_rate": 1.239142695995488e-05,
|
| 954 |
+
"loss": 0.0003,
|
| 955 |
+
"step": 1350
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 1.1505922165820643,
|
| 959 |
+
"grad_norm": 0.0031492102425545454,
|
| 960 |
+
"learning_rate": 1.233502538071066e-05,
|
| 961 |
+
"loss": 0.0003,
|
| 962 |
+
"step": 1360
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 1.1590524534686972,
|
| 966 |
+
"grad_norm": 0.0031071833800524473,
|
| 967 |
+
"learning_rate": 1.2278623801466443e-05,
|
| 968 |
+
"loss": 0.0006,
|
| 969 |
+
"step": 1370
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 1.16751269035533,
|
| 973 |
+
"grad_norm": 0.003029242157936096,
|
| 974 |
+
"learning_rate": 1.2222222222222224e-05,
|
| 975 |
+
"loss": 0.0004,
|
| 976 |
+
"step": 1380
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 1.1759729272419628,
|
| 980 |
+
"grad_norm": 0.003054770641028881,
|
| 981 |
+
"learning_rate": 1.2165820642978003e-05,
|
| 982 |
+
"loss": 0.0004,
|
| 983 |
+
"step": 1390
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 1.1844331641285957,
|
| 987 |
+
"grad_norm": 0.002985539613291621,
|
| 988 |
+
"learning_rate": 1.2109419063733784e-05,
|
| 989 |
+
"loss": 0.0003,
|
| 990 |
+
"step": 1400
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 1.1928934010152283,
|
| 994 |
+
"grad_norm": 0.0029861750081181526,
|
| 995 |
+
"learning_rate": 1.2053017484489567e-05,
|
| 996 |
+
"loss": 0.0003,
|
| 997 |
+
"step": 1410
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 1.2013536379018612,
|
| 1001 |
+
"grad_norm": 0.0031786304898560047,
|
| 1002 |
+
"learning_rate": 1.1996615905245348e-05,
|
| 1003 |
+
"loss": 0.0003,
|
| 1004 |
+
"step": 1420
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 1.2098138747884941,
|
| 1008 |
+
"grad_norm": 0.0029660591389983892,
|
| 1009 |
+
"learning_rate": 1.194021432600113e-05,
|
| 1010 |
+
"loss": 0.0003,
|
| 1011 |
+
"step": 1430
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 1.218274111675127,
|
| 1015 |
+
"grad_norm": 0.0029080845415592194,
|
| 1016 |
+
"learning_rate": 1.188381274675691e-05,
|
| 1017 |
+
"loss": 0.0003,
|
| 1018 |
+
"step": 1440
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 1.2267343485617597,
|
| 1022 |
+
"grad_norm": 0.003032978158444166,
|
| 1023 |
+
"learning_rate": 1.1827411167512693e-05,
|
| 1024 |
+
"loss": 0.0003,
|
| 1025 |
+
"step": 1450
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 1.2351945854483926,
|
| 1029 |
+
"grad_norm": 0.0028783201705664396,
|
| 1030 |
+
"learning_rate": 1.1771009588268472e-05,
|
| 1031 |
+
"loss": 0.0003,
|
| 1032 |
+
"step": 1460
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 1.2436548223350254,
|
| 1036 |
+
"grad_norm": 0.002919497899711132,
|
| 1037 |
+
"learning_rate": 1.1714608009024253e-05,
|
| 1038 |
+
"loss": 0.0003,
|
| 1039 |
+
"step": 1470
|
| 1040 |
+
},
|
| 1041 |
+
{
|
| 1042 |
+
"epoch": 1.252115059221658,
|
| 1043 |
+
"grad_norm": 0.002829916076734662,
|
| 1044 |
+
"learning_rate": 1.1658206429780034e-05,
|
| 1045 |
+
"loss": 0.0003,
|
| 1046 |
+
"step": 1480
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 1.260575296108291,
|
| 1050 |
+
"grad_norm": 0.0028766861651092768,
|
| 1051 |
+
"learning_rate": 1.1601804850535815e-05,
|
| 1052 |
+
"loss": 0.0003,
|
| 1053 |
+
"step": 1490
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 1.2690355329949239,
|
| 1057 |
+
"grad_norm": 0.0027290356811136007,
|
| 1058 |
+
"learning_rate": 1.1545403271291598e-05,
|
| 1059 |
+
"loss": 0.0003,
|
| 1060 |
+
"step": 1500
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 1.2774957698815568,
|
| 1064 |
+
"grad_norm": 0.0027461503632366657,
|
| 1065 |
+
"learning_rate": 1.1489001692047379e-05,
|
| 1066 |
+
"loss": 0.0003,
|
| 1067 |
+
"step": 1510
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 1.2859560067681894,
|
| 1071 |
+
"grad_norm": 0.0027623216155916452,
|
| 1072 |
+
"learning_rate": 1.143260011280316e-05,
|
| 1073 |
+
"loss": 0.0003,
|
| 1074 |
+
"step": 1520
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 1.2944162436548223,
|
| 1078 |
+
"grad_norm": 0.0028906099032610655,
|
| 1079 |
+
"learning_rate": 1.137619853355894e-05,
|
| 1080 |
+
"loss": 0.0003,
|
| 1081 |
+
"step": 1530
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 1.3028764805414552,
|
| 1085 |
+
"grad_norm": 0.0027789094019681215,
|
| 1086 |
+
"learning_rate": 1.1319796954314722e-05,
|
| 1087 |
+
"loss": 0.0003,
|
| 1088 |
+
"step": 1540
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 1.3113367174280879,
|
| 1092 |
+
"grad_norm": 0.00271011283621192,
|
| 1093 |
+
"learning_rate": 1.1263395375070503e-05,
|
| 1094 |
+
"loss": 0.0003,
|
| 1095 |
+
"step": 1550
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 1.3197969543147208,
|
| 1099 |
+
"grad_norm": 0.002672335831448436,
|
| 1100 |
+
"learning_rate": 1.1206993795826284e-05,
|
| 1101 |
+
"loss": 0.0003,
|
| 1102 |
+
"step": 1560
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 1.3282571912013537,
|
| 1106 |
+
"grad_norm": 0.002650258131325245,
|
| 1107 |
+
"learning_rate": 1.1150592216582065e-05,
|
| 1108 |
+
"loss": 0.0003,
|
| 1109 |
+
"step": 1570
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 1.3367174280879865,
|
| 1113 |
+
"grad_norm": 0.002780508017167449,
|
| 1114 |
+
"learning_rate": 1.1094190637337848e-05,
|
| 1115 |
+
"loss": 0.0003,
|
| 1116 |
+
"step": 1580
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 1.3451776649746192,
|
| 1120 |
+
"grad_norm": 0.0026137123350054026,
|
| 1121 |
+
"learning_rate": 1.1037789058093627e-05,
|
| 1122 |
+
"loss": 0.0003,
|
| 1123 |
+
"step": 1590
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 1.353637901861252,
|
| 1127 |
+
"grad_norm": 0.0026269517838954926,
|
| 1128 |
+
"learning_rate": 1.0981387478849408e-05,
|
| 1129 |
+
"loss": 0.0003,
|
| 1130 |
+
"step": 1600
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 1.362098138747885,
|
| 1134 |
+
"grad_norm": 0.0025848057121038437,
|
| 1135 |
+
"learning_rate": 1.092498589960519e-05,
|
| 1136 |
+
"loss": 0.0003,
|
| 1137 |
+
"step": 1610
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 1.3705583756345177,
|
| 1141 |
+
"grad_norm": 0.0025470599066466093,
|
| 1142 |
+
"learning_rate": 1.086858432036097e-05,
|
| 1143 |
+
"loss": 0.0003,
|
| 1144 |
+
"step": 1620
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 1.3790186125211505,
|
| 1148 |
+
"grad_norm": 0.0027439745608717203,
|
| 1149 |
+
"learning_rate": 1.0812182741116753e-05,
|
| 1150 |
+
"loss": 0.0003,
|
| 1151 |
+
"step": 1630
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 1.3874788494077834,
|
| 1155 |
+
"grad_norm": 0.002537579508498311,
|
| 1156 |
+
"learning_rate": 1.0755781161872534e-05,
|
| 1157 |
+
"loss": 0.0003,
|
| 1158 |
+
"step": 1640
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 1.3959390862944163,
|
| 1162 |
+
"grad_norm": 0.0025688863825052977,
|
| 1163 |
+
"learning_rate": 1.0699379582628315e-05,
|
| 1164 |
+
"loss": 0.0003,
|
| 1165 |
+
"step": 1650
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 1.404399323181049,
|
| 1169 |
+
"grad_norm": 0.0026179850101470947,
|
| 1170 |
+
"learning_rate": 1.0642978003384094e-05,
|
| 1171 |
+
"loss": 0.0003,
|
| 1172 |
+
"step": 1660
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 1.4128595600676819,
|
| 1176 |
+
"grad_norm": 0.002496235305443406,
|
| 1177 |
+
"learning_rate": 1.0586576424139875e-05,
|
| 1178 |
+
"loss": 0.0003,
|
| 1179 |
+
"step": 1670
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 1.4213197969543148,
|
| 1183 |
+
"grad_norm": 0.0026708145160228014,
|
| 1184 |
+
"learning_rate": 1.0530174844895658e-05,
|
| 1185 |
+
"loss": 0.0003,
|
| 1186 |
+
"step": 1680
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 1.4297800338409474,
|
| 1190 |
+
"grad_norm": 0.0025831579696387053,
|
| 1191 |
+
"learning_rate": 1.0473773265651439e-05,
|
| 1192 |
+
"loss": 0.0003,
|
| 1193 |
+
"step": 1690
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 1.4382402707275803,
|
| 1197 |
+
"grad_norm": 0.0024914774112403393,
|
| 1198 |
+
"learning_rate": 1.041737168640722e-05,
|
| 1199 |
+
"loss": 0.0003,
|
| 1200 |
+
"step": 1700
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 1.4467005076142132,
|
| 1204 |
+
"grad_norm": 0.002529192017391324,
|
| 1205 |
+
"learning_rate": 1.0360970107163003e-05,
|
| 1206 |
+
"loss": 0.0003,
|
| 1207 |
+
"step": 1710
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 1.455160744500846,
|
| 1211 |
+
"grad_norm": 0.0024393699131906033,
|
| 1212 |
+
"learning_rate": 1.0304568527918784e-05,
|
| 1213 |
+
"loss": 0.0011,
|
| 1214 |
+
"step": 1720
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 1.463620981387479,
|
| 1218 |
+
"grad_norm": 0.0043233660981059074,
|
| 1219 |
+
"learning_rate": 1.0248166948674563e-05,
|
| 1220 |
+
"loss": 0.0004,
|
| 1221 |
+
"step": 1730
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 1.4720812182741116,
|
| 1225 |
+
"grad_norm": 0.002547438722103834,
|
| 1226 |
+
"learning_rate": 1.0191765369430344e-05,
|
| 1227 |
+
"loss": 0.0005,
|
| 1228 |
+
"step": 1740
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 1.4805414551607445,
|
| 1232 |
+
"grad_norm": 0.0025178396608680487,
|
| 1233 |
+
"learning_rate": 1.0135363790186125e-05,
|
| 1234 |
+
"loss": 0.0003,
|
| 1235 |
+
"step": 1750
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 1.4890016920473772,
|
| 1239 |
+
"grad_norm": 0.003602534532546997,
|
| 1240 |
+
"learning_rate": 1.0078962210941908e-05,
|
| 1241 |
+
"loss": 0.0003,
|
| 1242 |
+
"step": 1760
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 1.49746192893401,
|
| 1246 |
+
"grad_norm": 0.002990358741953969,
|
| 1247 |
+
"learning_rate": 1.0022560631697689e-05,
|
| 1248 |
+
"loss": 0.0003,
|
| 1249 |
+
"step": 1770
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 1.505922165820643,
|
| 1253 |
+
"grad_norm": 0.0024538026191294193,
|
| 1254 |
+
"learning_rate": 9.96615905245347e-06,
|
| 1255 |
+
"loss": 0.0003,
|
| 1256 |
+
"step": 1780
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 1.5143824027072759,
|
| 1260 |
+
"grad_norm": 0.0024382509291172028,
|
| 1261 |
+
"learning_rate": 9.90975747320925e-06,
|
| 1262 |
+
"loss": 0.0003,
|
| 1263 |
+
"step": 1790
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 1.5228426395939088,
|
| 1267 |
+
"grad_norm": 0.0026386724784970284,
|
| 1268 |
+
"learning_rate": 9.853355893965032e-06,
|
| 1269 |
+
"loss": 0.0003,
|
| 1270 |
+
"step": 1800
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 1.5313028764805414,
|
| 1274 |
+
"grad_norm": 0.0023787450045347214,
|
| 1275 |
+
"learning_rate": 9.796954314720813e-06,
|
| 1276 |
+
"loss": 0.0003,
|
| 1277 |
+
"step": 1810
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 1.5397631133671743,
|
| 1281 |
+
"grad_norm": 0.002435258822515607,
|
| 1282 |
+
"learning_rate": 9.740552735476594e-06,
|
| 1283 |
+
"loss": 0.0003,
|
| 1284 |
+
"step": 1820
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 1.548223350253807,
|
| 1288 |
+
"grad_norm": 0.32252246141433716,
|
| 1289 |
+
"learning_rate": 9.684151156232375e-06,
|
| 1290 |
+
"loss": 0.0003,
|
| 1291 |
+
"step": 1830
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 1.5566835871404399,
|
| 1295 |
+
"grad_norm": 0.0023758108727633953,
|
| 1296 |
+
"learning_rate": 9.627749576988158e-06,
|
| 1297 |
+
"loss": 0.0003,
|
| 1298 |
+
"step": 1840
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 1.5651438240270727,
|
| 1302 |
+
"grad_norm": 0.0023211485240608454,
|
| 1303 |
+
"learning_rate": 9.571347997743937e-06,
|
| 1304 |
+
"loss": 0.0003,
|
| 1305 |
+
"step": 1850
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 1.5736040609137056,
|
| 1309 |
+
"grad_norm": 0.002314375713467598,
|
| 1310 |
+
"learning_rate": 9.514946418499718e-06,
|
| 1311 |
+
"loss": 0.0003,
|
| 1312 |
+
"step": 1860
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 1.5820642978003385,
|
| 1316 |
+
"grad_norm": 0.0023043151013553143,
|
| 1317 |
+
"learning_rate": 9.4585448392555e-06,
|
| 1318 |
+
"loss": 0.0003,
|
| 1319 |
+
"step": 1870
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 1.5905245346869712,
|
| 1323 |
+
"grad_norm": 0.0022787386551499367,
|
| 1324 |
+
"learning_rate": 9.402143260011282e-06,
|
| 1325 |
+
"loss": 0.0002,
|
| 1326 |
+
"step": 1880
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 1.598984771573604,
|
| 1330 |
+
"grad_norm": 0.0022805132903158665,
|
| 1331 |
+
"learning_rate": 9.345741680767063e-06,
|
| 1332 |
+
"loss": 0.0003,
|
| 1333 |
+
"step": 1890
|
| 1334 |
+
},
|
| 1335 |
+
{
|
| 1336 |
+
"epoch": 1.6074450084602367,
|
| 1337 |
+
"grad_norm": 0.0022434191778302193,
|
| 1338 |
+
"learning_rate": 9.289340101522844e-06,
|
| 1339 |
+
"loss": 0.0003,
|
| 1340 |
+
"step": 1900
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 1.6159052453468696,
|
| 1344 |
+
"grad_norm": 0.0022070412524044514,
|
| 1345 |
+
"learning_rate": 9.232938522278625e-06,
|
| 1346 |
+
"loss": 0.0002,
|
| 1347 |
+
"step": 1910
|
| 1348 |
+
},
|
| 1349 |
+
{
|
| 1350 |
+
"epoch": 1.6243654822335025,
|
| 1351 |
+
"grad_norm": 0.002298792591318488,
|
| 1352 |
+
"learning_rate": 9.176536943034406e-06,
|
| 1353 |
+
"loss": 0.0002,
|
| 1354 |
+
"step": 1920
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 1.6328257191201354,
|
| 1358 |
+
"grad_norm": 0.002347590634599328,
|
| 1359 |
+
"learning_rate": 9.120135363790187e-06,
|
| 1360 |
+
"loss": 0.0002,
|
| 1361 |
+
"step": 1930
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 1.6412859560067683,
|
| 1365 |
+
"grad_norm": 0.0022417185828089714,
|
| 1366 |
+
"learning_rate": 9.063733784545968e-06,
|
| 1367 |
+
"loss": 0.0002,
|
| 1368 |
+
"step": 1940
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 1.649746192893401,
|
| 1372 |
+
"grad_norm": 0.002178118098527193,
|
| 1373 |
+
"learning_rate": 9.007332205301749e-06,
|
| 1374 |
+
"loss": 0.0002,
|
| 1375 |
+
"step": 1950
|
| 1376 |
+
},
|
| 1377 |
+
{
|
| 1378 |
+
"epoch": 1.6582064297800339,
|
| 1379 |
+
"grad_norm": 0.002204876858741045,
|
| 1380 |
+
"learning_rate": 8.95093062605753e-06,
|
| 1381 |
+
"loss": 0.0002,
|
| 1382 |
+
"step": 1960
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 1.6666666666666665,
|
| 1386 |
+
"grad_norm": 0.0021794794593006372,
|
| 1387 |
+
"learning_rate": 8.894529046813312e-06,
|
| 1388 |
+
"loss": 0.0002,
|
| 1389 |
+
"step": 1970
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 1.6751269035532994,
|
| 1393 |
+
"grad_norm": 0.002448797458782792,
|
| 1394 |
+
"learning_rate": 8.838127467569093e-06,
|
| 1395 |
+
"loss": 0.0002,
|
| 1396 |
+
"step": 1980
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 1.6835871404399323,
|
| 1400 |
+
"grad_norm": 0.00216574571095407,
|
| 1401 |
+
"learning_rate": 8.781725888324873e-06,
|
| 1402 |
+
"loss": 0.0002,
|
| 1403 |
+
"step": 1990
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 1.6920473773265652,
|
| 1407 |
+
"grad_norm": 0.0022357029374688864,
|
| 1408 |
+
"learning_rate": 8.725324309080655e-06,
|
| 1409 |
+
"loss": 0.0002,
|
| 1410 |
+
"step": 2000
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"epoch": 1.700507614213198,
|
| 1414 |
+
"grad_norm": 0.0021086037158966064,
|
| 1415 |
+
"learning_rate": 8.668922729836436e-06,
|
| 1416 |
+
"loss": 0.0002,
|
| 1417 |
+
"step": 2010
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 1.708967851099831,
|
| 1421 |
+
"grad_norm": 0.002162441611289978,
|
| 1422 |
+
"learning_rate": 8.612521150592217e-06,
|
| 1423 |
+
"loss": 0.0002,
|
| 1424 |
+
"step": 2020
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 1.7174280879864636,
|
| 1428 |
+
"grad_norm": 0.0021641228813678026,
|
| 1429 |
+
"learning_rate": 8.556119571347998e-06,
|
| 1430 |
+
"loss": 0.0011,
|
| 1431 |
+
"step": 2030
|
| 1432 |
+
},
|
| 1433 |
+
{
|
| 1434 |
+
"epoch": 1.7258883248730963,
|
| 1435 |
+
"grad_norm": 0.0021566457580775023,
|
| 1436 |
+
"learning_rate": 8.49971799210378e-06,
|
| 1437 |
+
"loss": 0.0002,
|
| 1438 |
+
"step": 2040
|
| 1439 |
+
},
|
| 1440 |
+
{
|
| 1441 |
+
"epoch": 1.7343485617597292,
|
| 1442 |
+
"grad_norm": 0.0021678991615772247,
|
| 1443 |
+
"learning_rate": 8.44331641285956e-06,
|
| 1444 |
+
"loss": 0.0002,
|
| 1445 |
+
"step": 2050
|
| 1446 |
+
},
|
| 1447 |
+
{
|
| 1448 |
+
"epoch": 1.742808798646362,
|
| 1449 |
+
"grad_norm": 0.002119549084454775,
|
| 1450 |
+
"learning_rate": 8.386914833615342e-06,
|
| 1451 |
+
"loss": 0.0002,
|
| 1452 |
+
"step": 2060
|
| 1453 |
+
},
|
| 1454 |
+
{
|
| 1455 |
+
"epoch": 1.751269035532995,
|
| 1456 |
+
"grad_norm": 0.002093096962198615,
|
| 1457 |
+
"learning_rate": 8.330513254371123e-06,
|
| 1458 |
+
"loss": 0.0002,
|
| 1459 |
+
"step": 2070
|
| 1460 |
+
},
|
| 1461 |
+
{
|
| 1462 |
+
"epoch": 1.7597292724196278,
|
| 1463 |
+
"grad_norm": 0.002090982161462307,
|
| 1464 |
+
"learning_rate": 8.274111675126905e-06,
|
| 1465 |
+
"loss": 0.0002,
|
| 1466 |
+
"step": 2080
|
| 1467 |
+
},
|
| 1468 |
+
{
|
| 1469 |
+
"epoch": 1.7681895093062607,
|
| 1470 |
+
"grad_norm": 0.004076903220266104,
|
| 1471 |
+
"learning_rate": 8.217710095882685e-06,
|
| 1472 |
+
"loss": 0.0002,
|
| 1473 |
+
"step": 2090
|
| 1474 |
+
},
|
| 1475 |
+
{
|
| 1476 |
+
"epoch": 1.7766497461928934,
|
| 1477 |
+
"grad_norm": 0.0020308117382228374,
|
| 1478 |
+
"learning_rate": 8.161308516638467e-06,
|
| 1479 |
+
"loss": 0.0002,
|
| 1480 |
+
"step": 2100
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"epoch": 1.785109983079526,
|
| 1484 |
+
"grad_norm": 0.0021146961953490973,
|
| 1485 |
+
"learning_rate": 8.104906937394248e-06,
|
| 1486 |
+
"loss": 0.0002,
|
| 1487 |
+
"step": 2110
|
| 1488 |
+
},
|
| 1489 |
+
{
|
| 1490 |
+
"epoch": 1.793570219966159,
|
| 1491 |
+
"grad_norm": 0.0020670590456575155,
|
| 1492 |
+
"learning_rate": 8.048505358150028e-06,
|
| 1493 |
+
"loss": 0.0002,
|
| 1494 |
+
"step": 2120
|
| 1495 |
+
},
|
| 1496 |
+
{
|
| 1497 |
+
"epoch": 1.8020304568527918,
|
| 1498 |
+
"grad_norm": 0.0024381864350289106,
|
| 1499 |
+
"learning_rate": 7.99210377890581e-06,
|
| 1500 |
+
"loss": 0.0002,
|
| 1501 |
+
"step": 2130
|
| 1502 |
+
},
|
| 1503 |
+
{
|
| 1504 |
+
"epoch": 1.8104906937394247,
|
| 1505 |
+
"grad_norm": 0.002149208215996623,
|
| 1506 |
+
"learning_rate": 7.935702199661591e-06,
|
| 1507 |
+
"loss": 0.0008,
|
| 1508 |
+
"step": 2140
|
| 1509 |
+
},
|
| 1510 |
+
{
|
| 1511 |
+
"epoch": 1.8189509306260576,
|
| 1512 |
+
"grad_norm": 0.004185215570032597,
|
| 1513 |
+
"learning_rate": 7.879300620417372e-06,
|
| 1514 |
+
"loss": 0.0002,
|
| 1515 |
+
"step": 2150
|
| 1516 |
+
},
|
| 1517 |
+
{
|
| 1518 |
+
"epoch": 1.8274111675126905,
|
| 1519 |
+
"grad_norm": 0.002098744735121727,
|
| 1520 |
+
"learning_rate": 7.822899041173153e-06,
|
| 1521 |
+
"loss": 0.0002,
|
| 1522 |
+
"step": 2160
|
| 1523 |
+
},
|
| 1524 |
+
{
|
| 1525 |
+
"epoch": 1.8358714043993232,
|
| 1526 |
+
"grad_norm": 0.0019923567306250334,
|
| 1527 |
+
"learning_rate": 7.766497461928934e-06,
|
| 1528 |
+
"loss": 0.0004,
|
| 1529 |
+
"step": 2170
|
| 1530 |
+
},
|
| 1531 |
+
{
|
| 1532 |
+
"epoch": 1.844331641285956,
|
| 1533 |
+
"grad_norm": 0.002045057248324156,
|
| 1534 |
+
"learning_rate": 7.710095882684715e-06,
|
| 1535 |
+
"loss": 0.0002,
|
| 1536 |
+
"step": 2180
|
| 1537 |
+
},
|
| 1538 |
+
{
|
| 1539 |
+
"epoch": 1.8527918781725887,
|
| 1540 |
+
"grad_norm": 0.002222336595878005,
|
| 1541 |
+
"learning_rate": 7.653694303440496e-06,
|
| 1542 |
+
"loss": 0.0002,
|
| 1543 |
+
"step": 2190
|
| 1544 |
+
},
|
| 1545 |
+
{
|
| 1546 |
+
"epoch": 1.8612521150592216,
|
| 1547 |
+
"grad_norm": 0.0021004409063607454,
|
| 1548 |
+
"learning_rate": 7.597292724196278e-06,
|
| 1549 |
+
"loss": 0.0002,
|
| 1550 |
+
"step": 2200
|
| 1551 |
+
},
|
| 1552 |
+
{
|
| 1553 |
+
"epoch": 1.8697123519458545,
|
| 1554 |
+
"grad_norm": 0.001979230437427759,
|
| 1555 |
+
"learning_rate": 7.540891144952059e-06,
|
| 1556 |
+
"loss": 0.0002,
|
| 1557 |
+
"step": 2210
|
| 1558 |
+
},
|
| 1559 |
+
{
|
| 1560 |
+
"epoch": 1.8781725888324874,
|
| 1561 |
+
"grad_norm": 0.0020105023868381977,
|
| 1562 |
+
"learning_rate": 7.484489565707841e-06,
|
| 1563 |
+
"loss": 0.0002,
|
| 1564 |
+
"step": 2220
|
| 1565 |
+
},
|
| 1566 |
+
{
|
| 1567 |
+
"epoch": 1.8866328257191203,
|
| 1568 |
+
"grad_norm": 0.0020976634696125984,
|
| 1569 |
+
"learning_rate": 7.428087986463621e-06,
|
| 1570 |
+
"loss": 0.0002,
|
| 1571 |
+
"step": 2230
|
| 1572 |
+
},
|
| 1573 |
+
{
|
| 1574 |
+
"epoch": 1.895093062605753,
|
| 1575 |
+
"grad_norm": 0.0020316061563789845,
|
| 1576 |
+
"learning_rate": 7.371686407219403e-06,
|
| 1577 |
+
"loss": 0.0002,
|
| 1578 |
+
"step": 2240
|
| 1579 |
+
},
|
| 1580 |
+
{
|
| 1581 |
+
"epoch": 1.9035532994923858,
|
| 1582 |
+
"grad_norm": 0.00200808048248291,
|
| 1583 |
+
"learning_rate": 7.315284827975184e-06,
|
| 1584 |
+
"loss": 0.0002,
|
| 1585 |
+
"step": 2250
|
| 1586 |
+
},
|
| 1587 |
+
{
|
| 1588 |
+
"epoch": 1.9120135363790185,
|
| 1589 |
+
"grad_norm": 0.002250692341476679,
|
| 1590 |
+
"learning_rate": 7.258883248730964e-06,
|
| 1591 |
+
"loss": 0.0002,
|
| 1592 |
+
"step": 2260
|
| 1593 |
+
},
|
| 1594 |
+
{
|
| 1595 |
+
"epoch": 1.9204737732656514,
|
| 1596 |
+
"grad_norm": 0.0022177433129400015,
|
| 1597 |
+
"learning_rate": 7.202481669486746e-06,
|
| 1598 |
+
"loss": 0.0007,
|
| 1599 |
+
"step": 2270
|
| 1600 |
+
},
|
| 1601 |
+
{
|
| 1602 |
+
"epoch": 1.9289340101522843,
|
| 1603 |
+
"grad_norm": 0.0018741836538538337,
|
| 1604 |
+
"learning_rate": 7.146080090242527e-06,
|
| 1605 |
+
"loss": 0.0002,
|
| 1606 |
+
"step": 2280
|
| 1607 |
+
},
|
| 1608 |
+
{
|
| 1609 |
+
"epoch": 1.9373942470389172,
|
| 1610 |
+
"grad_norm": 0.002023757202550769,
|
| 1611 |
+
"learning_rate": 7.089678510998309e-06,
|
| 1612 |
+
"loss": 0.0002,
|
| 1613 |
+
"step": 2290
|
| 1614 |
+
},
|
| 1615 |
+
{
|
| 1616 |
+
"epoch": 1.94585448392555,
|
| 1617 |
+
"grad_norm": 0.0021521553862839937,
|
| 1618 |
+
"learning_rate": 7.033276931754089e-06,
|
| 1619 |
+
"loss": 0.0002,
|
| 1620 |
+
"step": 2300
|
| 1621 |
+
},
|
| 1622 |
+
{
|
| 1623 |
+
"epoch": 1.9543147208121827,
|
| 1624 |
+
"grad_norm": 0.0019342171726748347,
|
| 1625 |
+
"learning_rate": 6.976875352509871e-06,
|
| 1626 |
+
"loss": 0.0008,
|
| 1627 |
+
"step": 2310
|
| 1628 |
+
},
|
| 1629 |
+
{
|
| 1630 |
+
"epoch": 1.9627749576988156,
|
| 1631 |
+
"grad_norm": 0.0019481374183669686,
|
| 1632 |
+
"learning_rate": 6.920473773265652e-06,
|
| 1633 |
+
"loss": 0.0007,
|
| 1634 |
+
"step": 2320
|
| 1635 |
+
},
|
| 1636 |
+
{
|
| 1637 |
+
"epoch": 1.9712351945854483,
|
| 1638 |
+
"grad_norm": 0.002077361335977912,
|
| 1639 |
+
"learning_rate": 6.864072194021433e-06,
|
| 1640 |
+
"loss": 0.0011,
|
| 1641 |
+
"step": 2330
|
| 1642 |
+
},
|
| 1643 |
+
{
|
| 1644 |
+
"epoch": 1.9796954314720812,
|
| 1645 |
+
"grad_norm": 0.0020031894091516733,
|
| 1646 |
+
"learning_rate": 6.807670614777214e-06,
|
| 1647 |
+
"loss": 0.0003,
|
| 1648 |
+
"step": 2340
|
| 1649 |
+
},
|
| 1650 |
+
{
|
| 1651 |
+
"epoch": 1.988155668358714,
|
| 1652 |
+
"grad_norm": 0.00219974503852427,
|
| 1653 |
+
"learning_rate": 6.751269035532996e-06,
|
| 1654 |
+
"loss": 0.0002,
|
| 1655 |
+
"step": 2350
|
| 1656 |
+
},
|
| 1657 |
+
{
|
| 1658 |
+
"epoch": 1.996615905245347,
|
| 1659 |
+
"grad_norm": 0.002148480387404561,
|
| 1660 |
+
"learning_rate": 6.694867456288776e-06,
|
| 1661 |
+
"loss": 0.0002,
|
| 1662 |
+
"step": 2360
|
| 1663 |
+
},
|
| 1664 |
+
{
|
| 1665 |
+
"epoch": 2.00507614213198,
|
| 1666 |
+
"grad_norm": 0.0018948954530060291,
|
| 1667 |
+
"learning_rate": 6.638465877044558e-06,
|
| 1668 |
+
"loss": 0.0002,
|
| 1669 |
+
"step": 2370
|
| 1670 |
+
},
|
| 1671 |
+
{
|
| 1672 |
+
"epoch": 2.0135363790186127,
|
| 1673 |
+
"grad_norm": 0.007534192409366369,
|
| 1674 |
+
"learning_rate": 6.582064297800339e-06,
|
| 1675 |
+
"loss": 0.0002,
|
| 1676 |
+
"step": 2380
|
| 1677 |
+
},
|
| 1678 |
+
{
|
| 1679 |
+
"epoch": 2.021996615905245,
|
| 1680 |
+
"grad_norm": 0.0018520756857469678,
|
| 1681 |
+
"learning_rate": 6.525662718556121e-06,
|
| 1682 |
+
"loss": 0.0002,
|
| 1683 |
+
"step": 2390
|
| 1684 |
+
},
|
| 1685 |
+
{
|
| 1686 |
+
"epoch": 2.030456852791878,
|
| 1687 |
+
"grad_norm": 0.0020236349664628506,
|
| 1688 |
+
"learning_rate": 6.469261139311901e-06,
|
| 1689 |
+
"loss": 0.0002,
|
| 1690 |
+
"step": 2400
|
| 1691 |
+
},
|
| 1692 |
+
{
|
| 1693 |
+
"epoch": 2.038917089678511,
|
| 1694 |
+
"grad_norm": 0.0018428469775244594,
|
| 1695 |
+
"learning_rate": 6.412859560067682e-06,
|
| 1696 |
+
"loss": 0.0002,
|
| 1697 |
+
"step": 2410
|
| 1698 |
+
},
|
| 1699 |
+
{
|
| 1700 |
+
"epoch": 2.047377326565144,
|
| 1701 |
+
"grad_norm": 0.0018592760898172855,
|
| 1702 |
+
"learning_rate": 6.356457980823464e-06,
|
| 1703 |
+
"loss": 0.0002,
|
| 1704 |
+
"step": 2420
|
| 1705 |
+
},
|
| 1706 |
+
{
|
| 1707 |
+
"epoch": 2.0558375634517767,
|
| 1708 |
+
"grad_norm": 0.001955309882760048,
|
| 1709 |
+
"learning_rate": 6.300056401579244e-06,
|
| 1710 |
+
"loss": 0.0002,
|
| 1711 |
+
"step": 2430
|
| 1712 |
+
},
|
| 1713 |
+
{
|
| 1714 |
+
"epoch": 2.0642978003384096,
|
| 1715 |
+
"grad_norm": 0.001861874246969819,
|
| 1716 |
+
"learning_rate": 6.243654822335026e-06,
|
| 1717 |
+
"loss": 0.0002,
|
| 1718 |
+
"step": 2440
|
| 1719 |
+
},
|
| 1720 |
+
{
|
| 1721 |
+
"epoch": 2.0727580372250425,
|
| 1722 |
+
"grad_norm": 0.0018448787741363049,
|
| 1723 |
+
"learning_rate": 6.187253243090807e-06,
|
| 1724 |
+
"loss": 0.0002,
|
| 1725 |
+
"step": 2450
|
| 1726 |
+
},
|
| 1727 |
+
{
|
| 1728 |
+
"epoch": 2.081218274111675,
|
| 1729 |
+
"grad_norm": 0.0018381918780505657,
|
| 1730 |
+
"learning_rate": 6.130851663846589e-06,
|
| 1731 |
+
"loss": 0.0002,
|
| 1732 |
+
"step": 2460
|
| 1733 |
+
},
|
| 1734 |
+
{
|
| 1735 |
+
"epoch": 2.089678510998308,
|
| 1736 |
+
"grad_norm": 0.0018074287800118327,
|
| 1737 |
+
"learning_rate": 6.074450084602369e-06,
|
| 1738 |
+
"loss": 0.0002,
|
| 1739 |
+
"step": 2470
|
| 1740 |
+
},
|
| 1741 |
+
{
|
| 1742 |
+
"epoch": 2.0981387478849407,
|
| 1743 |
+
"grad_norm": 0.0018625473603606224,
|
| 1744 |
+
"learning_rate": 6.018048505358151e-06,
|
| 1745 |
+
"loss": 0.0002,
|
| 1746 |
+
"step": 2480
|
| 1747 |
+
},
|
| 1748 |
+
{
|
| 1749 |
+
"epoch": 2.1065989847715736,
|
| 1750 |
+
"grad_norm": 0.0017794481245800853,
|
| 1751 |
+
"learning_rate": 5.961646926113932e-06,
|
| 1752 |
+
"loss": 0.0002,
|
| 1753 |
+
"step": 2490
|
| 1754 |
+
},
|
| 1755 |
+
{
|
| 1756 |
+
"epoch": 2.1150592216582065,
|
| 1757 |
+
"grad_norm": 0.0017721441108733416,
|
| 1758 |
+
"learning_rate": 5.905245346869712e-06,
|
| 1759 |
+
"loss": 0.0002,
|
| 1760 |
+
"step": 2500
|
| 1761 |
+
},
|
| 1762 |
+
{
|
| 1763 |
+
"epoch": 2.1235194585448394,
|
| 1764 |
+
"grad_norm": 0.005279191303998232,
|
| 1765 |
+
"learning_rate": 5.848843767625494e-06,
|
| 1766 |
+
"loss": 0.0002,
|
| 1767 |
+
"step": 2510
|
| 1768 |
+
},
|
| 1769 |
+
{
|
| 1770 |
+
"epoch": 2.1319796954314723,
|
| 1771 |
+
"grad_norm": 0.0017379368655383587,
|
| 1772 |
+
"learning_rate": 5.792442188381276e-06,
|
| 1773 |
+
"loss": 0.0002,
|
| 1774 |
+
"step": 2520
|
| 1775 |
+
},
|
| 1776 |
+
{
|
| 1777 |
+
"epoch": 2.1404399323181047,
|
| 1778 |
+
"grad_norm": 0.001751760020852089,
|
| 1779 |
+
"learning_rate": 5.736040609137057e-06,
|
| 1780 |
+
"loss": 0.0002,
|
| 1781 |
+
"step": 2530
|
| 1782 |
+
},
|
| 1783 |
+
{
|
| 1784 |
+
"epoch": 2.1489001692047376,
|
| 1785 |
+
"grad_norm": 0.001752077485434711,
|
| 1786 |
+
"learning_rate": 5.679639029892837e-06,
|
| 1787 |
+
"loss": 0.0002,
|
| 1788 |
+
"step": 2540
|
| 1789 |
+
},
|
| 1790 |
+
{
|
| 1791 |
+
"epoch": 2.1573604060913705,
|
| 1792 |
+
"grad_norm": 0.0017231889069080353,
|
| 1793 |
+
"learning_rate": 5.623237450648619e-06,
|
| 1794 |
+
"loss": 0.0002,
|
| 1795 |
+
"step": 2550
|
| 1796 |
+
},
|
| 1797 |
+
{
|
| 1798 |
+
"epoch": 2.1658206429780034,
|
| 1799 |
+
"grad_norm": 0.0017523319693282247,
|
| 1800 |
+
"learning_rate": 5.5668358714044e-06,
|
| 1801 |
+
"loss": 0.0002,
|
| 1802 |
+
"step": 2560
|
| 1803 |
+
},
|
| 1804 |
+
{
|
| 1805 |
+
"epoch": 2.1742808798646363,
|
| 1806 |
+
"grad_norm": 0.0017014428740367293,
|
| 1807 |
+
"learning_rate": 5.510434292160181e-06,
|
| 1808 |
+
"loss": 0.0002,
|
| 1809 |
+
"step": 2570
|
| 1810 |
+
},
|
| 1811 |
+
{
|
| 1812 |
+
"epoch": 2.182741116751269,
|
| 1813 |
+
"grad_norm": 0.0017529787728562951,
|
| 1814 |
+
"learning_rate": 5.454032712915962e-06,
|
| 1815 |
+
"loss": 0.0002,
|
| 1816 |
+
"step": 2580
|
| 1817 |
+
},
|
| 1818 |
+
{
|
| 1819 |
+
"epoch": 2.191201353637902,
|
| 1820 |
+
"grad_norm": 0.0017768230754882097,
|
| 1821 |
+
"learning_rate": 5.397631133671744e-06,
|
| 1822 |
+
"loss": 0.0002,
|
| 1823 |
+
"step": 2590
|
| 1824 |
+
},
|
| 1825 |
+
{
|
| 1826 |
+
"epoch": 2.199661590524535,
|
| 1827 |
+
"grad_norm": 0.0017734098946675658,
|
| 1828 |
+
"learning_rate": 5.341229554427524e-06,
|
| 1829 |
+
"loss": 0.0002,
|
| 1830 |
+
"step": 2600
|
| 1831 |
+
},
|
| 1832 |
+
{
|
| 1833 |
+
"epoch": 2.2081218274111674,
|
| 1834 |
+
"grad_norm": 0.00168097042478621,
|
| 1835 |
+
"learning_rate": 5.284827975183306e-06,
|
| 1836 |
+
"loss": 0.0002,
|
| 1837 |
+
"step": 2610
|
| 1838 |
+
},
|
| 1839 |
+
{
|
| 1840 |
+
"epoch": 2.2165820642978002,
|
| 1841 |
+
"grad_norm": 0.0017450157320126891,
|
| 1842 |
+
"learning_rate": 5.228426395939087e-06,
|
| 1843 |
+
"loss": 0.0002,
|
| 1844 |
+
"step": 2620
|
| 1845 |
+
},
|
| 1846 |
+
{
|
| 1847 |
+
"epoch": 2.225042301184433,
|
| 1848 |
+
"grad_norm": 0.0016924841329455376,
|
| 1849 |
+
"learning_rate": 5.1720248166948686e-06,
|
| 1850 |
+
"loss": 0.0002,
|
| 1851 |
+
"step": 2630
|
| 1852 |
+
},
|
| 1853 |
+
{
|
| 1854 |
+
"epoch": 2.233502538071066,
|
| 1855 |
+
"grad_norm": 0.0017653257818892598,
|
| 1856 |
+
"learning_rate": 5.115623237450649e-06,
|
| 1857 |
+
"loss": 0.0002,
|
| 1858 |
+
"step": 2640
|
| 1859 |
+
},
|
| 1860 |
+
{
|
| 1861 |
+
"epoch": 2.241962774957699,
|
| 1862 |
+
"grad_norm": 0.00172147445846349,
|
| 1863 |
+
"learning_rate": 5.059221658206431e-06,
|
| 1864 |
+
"loss": 0.0002,
|
| 1865 |
+
"step": 2650
|
| 1866 |
+
},
|
| 1867 |
+
{
|
| 1868 |
+
"epoch": 2.250423011844332,
|
| 1869 |
+
"grad_norm": 0.0017959652468562126,
|
| 1870 |
+
"learning_rate": 5.002820078962212e-06,
|
| 1871 |
+
"loss": 0.0002,
|
| 1872 |
+
"step": 2660
|
| 1873 |
+
},
|
| 1874 |
+
{
|
| 1875 |
+
"epoch": 2.2588832487309647,
|
| 1876 |
+
"grad_norm": 0.0016738857375457883,
|
| 1877 |
+
"learning_rate": 4.946418499717993e-06,
|
| 1878 |
+
"loss": 0.0002,
|
| 1879 |
+
"step": 2670
|
| 1880 |
+
},
|
| 1881 |
+
{
|
| 1882 |
+
"epoch": 2.267343485617597,
|
| 1883 |
+
"grad_norm": 0.0016783688915893435,
|
| 1884 |
+
"learning_rate": 4.890016920473774e-06,
|
| 1885 |
+
"loss": 0.0002,
|
| 1886 |
+
"step": 2680
|
| 1887 |
+
},
|
| 1888 |
+
{
|
| 1889 |
+
"epoch": 2.27580372250423,
|
| 1890 |
+
"grad_norm": 0.001678799162618816,
|
| 1891 |
+
"learning_rate": 4.833615341229555e-06,
|
| 1892 |
+
"loss": 0.0002,
|
| 1893 |
+
"step": 2690
|
| 1894 |
+
},
|
| 1895 |
+
{
|
| 1896 |
+
"epoch": 2.284263959390863,
|
| 1897 |
+
"grad_norm": 0.0017365720123052597,
|
| 1898 |
+
"learning_rate": 4.777213761985336e-06,
|
| 1899 |
+
"loss": 0.0002,
|
| 1900 |
+
"step": 2700
|
| 1901 |
+
},
|
| 1902 |
+
{
|
| 1903 |
+
"epoch": 2.292724196277496,
|
| 1904 |
+
"grad_norm": 0.00171087181661278,
|
| 1905 |
+
"learning_rate": 4.7208121827411175e-06,
|
| 1906 |
+
"loss": 0.0002,
|
| 1907 |
+
"step": 2710
|
| 1908 |
+
},
|
| 1909 |
+
{
|
| 1910 |
+
"epoch": 2.3011844331641287,
|
| 1911 |
+
"grad_norm": 0.001663476345129311,
|
| 1912 |
+
"learning_rate": 4.6644106034968985e-06,
|
| 1913 |
+
"loss": 0.0002,
|
| 1914 |
+
"step": 2720
|
| 1915 |
+
},
|
| 1916 |
+
{
|
| 1917 |
+
"epoch": 2.3096446700507616,
|
| 1918 |
+
"grad_norm": 0.0016244918806478381,
|
| 1919 |
+
"learning_rate": 4.6080090242526796e-06,
|
| 1920 |
+
"loss": 0.0002,
|
| 1921 |
+
"step": 2730
|
| 1922 |
+
},
|
| 1923 |
+
{
|
| 1924 |
+
"epoch": 2.3181049069373945,
|
| 1925 |
+
"grad_norm": 0.0016214584466069937,
|
| 1926 |
+
"learning_rate": 4.5516074450084606e-06,
|
| 1927 |
+
"loss": 0.0002,
|
| 1928 |
+
"step": 2740
|
| 1929 |
+
},
|
| 1930 |
+
{
|
| 1931 |
+
"epoch": 2.326565143824027,
|
| 1932 |
+
"grad_norm": 0.0016812806716188788,
|
| 1933 |
+
"learning_rate": 4.4952058657642416e-06,
|
| 1934 |
+
"loss": 0.0002,
|
| 1935 |
+
"step": 2750
|
| 1936 |
+
},
|
| 1937 |
+
{
|
| 1938 |
+
"epoch": 2.33502538071066,
|
| 1939 |
+
"grad_norm": 0.0016772805247455835,
|
| 1940 |
+
"learning_rate": 4.4388042865200234e-06,
|
| 1941 |
+
"loss": 0.0002,
|
| 1942 |
+
"step": 2760
|
| 1943 |
+
},
|
| 1944 |
+
{
|
| 1945 |
+
"epoch": 2.3434856175972927,
|
| 1946 |
+
"grad_norm": 0.0016064831288531423,
|
| 1947 |
+
"learning_rate": 4.3824027072758044e-06,
|
| 1948 |
+
"loss": 0.0002,
|
| 1949 |
+
"step": 2770
|
| 1950 |
+
},
|
| 1951 |
+
{
|
| 1952 |
+
"epoch": 2.3519458544839256,
|
| 1953 |
+
"grad_norm": 0.0016283520963042974,
|
| 1954 |
+
"learning_rate": 4.326001128031585e-06,
|
| 1955 |
+
"loss": 0.0002,
|
| 1956 |
+
"step": 2780
|
| 1957 |
+
},
|
| 1958 |
+
{
|
| 1959 |
+
"epoch": 2.3604060913705585,
|
| 1960 |
+
"grad_norm": 0.0017511517507955432,
|
| 1961 |
+
"learning_rate": 4.2695995487873665e-06,
|
| 1962 |
+
"loss": 0.0002,
|
| 1963 |
+
"step": 2790
|
| 1964 |
+
},
|
| 1965 |
+
{
|
| 1966 |
+
"epoch": 2.3688663282571913,
|
| 1967 |
+
"grad_norm": 0.001622919924557209,
|
| 1968 |
+
"learning_rate": 4.2131979695431475e-06,
|
| 1969 |
+
"loss": 0.0002,
|
| 1970 |
+
"step": 2800
|
| 1971 |
+
},
|
| 1972 |
+
{
|
| 1973 |
+
"epoch": 2.3773265651438242,
|
| 1974 |
+
"grad_norm": 0.0017815829487517476,
|
| 1975 |
+
"learning_rate": 4.1567963902989285e-06,
|
| 1976 |
+
"loss": 0.0004,
|
| 1977 |
+
"step": 2810
|
| 1978 |
+
},
|
| 1979 |
+
{
|
| 1980 |
+
"epoch": 2.3857868020304567,
|
| 1981 |
+
"grad_norm": 0.001768069458194077,
|
| 1982 |
+
"learning_rate": 4.1003948110547095e-06,
|
| 1983 |
+
"loss": 0.0002,
|
| 1984 |
+
"step": 2820
|
| 1985 |
+
},
|
| 1986 |
+
{
|
| 1987 |
+
"epoch": 2.3942470389170896,
|
| 1988 |
+
"grad_norm": 0.001824942184612155,
|
| 1989 |
+
"learning_rate": 4.0439932318104905e-06,
|
| 1990 |
+
"loss": 0.0002,
|
| 1991 |
+
"step": 2830
|
| 1992 |
+
},
|
| 1993 |
+
{
|
| 1994 |
+
"epoch": 2.4027072758037225,
|
| 1995 |
+
"grad_norm": 0.0016344970790669322,
|
| 1996 |
+
"learning_rate": 3.987591652566272e-06,
|
| 1997 |
+
"loss": 0.0005,
|
| 1998 |
+
"step": 2840
|
| 1999 |
+
},
|
| 2000 |
+
{
|
| 2001 |
+
"epoch": 2.4111675126903553,
|
| 2002 |
+
"grad_norm": 0.0016487556276842952,
|
| 2003 |
+
"learning_rate": 3.931190073322053e-06,
|
| 2004 |
+
"loss": 0.0002,
|
| 2005 |
+
"step": 2850
|
| 2006 |
+
},
|
| 2007 |
+
{
|
| 2008 |
+
"epoch": 2.4196277495769882,
|
| 2009 |
+
"grad_norm": 0.0016864290228113532,
|
| 2010 |
+
"learning_rate": 3.874788494077834e-06,
|
| 2011 |
+
"loss": 0.0002,
|
| 2012 |
+
"step": 2860
|
| 2013 |
+
},
|
| 2014 |
+
{
|
| 2015 |
+
"epoch": 2.428087986463621,
|
| 2016 |
+
"grad_norm": 0.0017508761957287788,
|
| 2017 |
+
"learning_rate": 3.8183869148336154e-06,
|
| 2018 |
+
"loss": 0.0002,
|
| 2019 |
+
"step": 2870
|
| 2020 |
+
},
|
| 2021 |
+
{
|
| 2022 |
+
"epoch": 2.436548223350254,
|
| 2023 |
+
"grad_norm": 0.0016393917612731457,
|
| 2024 |
+
"learning_rate": 3.761985335589397e-06,
|
| 2025 |
+
"loss": 0.0002,
|
| 2026 |
+
"step": 2880
|
| 2027 |
+
},
|
| 2028 |
+
{
|
| 2029 |
+
"epoch": 2.4450084602368864,
|
| 2030 |
+
"grad_norm": 0.001650340622290969,
|
| 2031 |
+
"learning_rate": 3.705583756345178e-06,
|
| 2032 |
+
"loss": 0.0002,
|
| 2033 |
+
"step": 2890
|
| 2034 |
+
},
|
| 2035 |
+
{
|
| 2036 |
+
"epoch": 2.4534686971235193,
|
| 2037 |
+
"grad_norm": 0.0016186356078833342,
|
| 2038 |
+
"learning_rate": 3.649182177100959e-06,
|
| 2039 |
+
"loss": 0.0002,
|
| 2040 |
+
"step": 2900
|
| 2041 |
+
},
|
| 2042 |
+
{
|
| 2043 |
+
"epoch": 2.4619289340101522,
|
| 2044 |
+
"grad_norm": 0.001632276806049049,
|
| 2045 |
+
"learning_rate": 3.5927805978567403e-06,
|
| 2046 |
+
"loss": 0.0002,
|
| 2047 |
+
"step": 2910
|
| 2048 |
+
},
|
| 2049 |
+
{
|
| 2050 |
+
"epoch": 2.470389170896785,
|
| 2051 |
+
"grad_norm": 0.0017024771077558398,
|
| 2052 |
+
"learning_rate": 3.5363790186125213e-06,
|
| 2053 |
+
"loss": 0.0002,
|
| 2054 |
+
"step": 2920
|
| 2055 |
+
},
|
| 2056 |
+
{
|
| 2057 |
+
"epoch": 2.478849407783418,
|
| 2058 |
+
"grad_norm": 0.0015938036376610398,
|
| 2059 |
+
"learning_rate": 3.4799774393683028e-06,
|
| 2060 |
+
"loss": 0.0002,
|
| 2061 |
+
"step": 2930
|
| 2062 |
+
},
|
| 2063 |
+
{
|
| 2064 |
+
"epoch": 2.487309644670051,
|
| 2065 |
+
"grad_norm": 0.001573930960148573,
|
| 2066 |
+
"learning_rate": 3.4235758601240838e-06,
|
| 2067 |
+
"loss": 0.0002,
|
| 2068 |
+
"step": 2940
|
| 2069 |
+
},
|
| 2070 |
+
{
|
| 2071 |
+
"epoch": 2.495769881556684,
|
| 2072 |
+
"grad_norm": 0.0015760001260787249,
|
| 2073 |
+
"learning_rate": 3.367174280879865e-06,
|
| 2074 |
+
"loss": 0.0002,
|
| 2075 |
+
"step": 2950
|
| 2076 |
+
},
|
| 2077 |
+
{
|
| 2078 |
+
"epoch": 2.504230118443316,
|
| 2079 |
+
"grad_norm": 0.0016050884732976556,
|
| 2080 |
+
"learning_rate": 3.3107727016356462e-06,
|
| 2081 |
+
"loss": 0.0002,
|
| 2082 |
+
"step": 2960
|
| 2083 |
+
},
|
| 2084 |
+
{
|
| 2085 |
+
"epoch": 2.512690355329949,
|
| 2086 |
+
"grad_norm": 0.0017449154984205961,
|
| 2087 |
+
"learning_rate": 3.2543711223914273e-06,
|
| 2088 |
+
"loss": 0.0002,
|
| 2089 |
+
"step": 2970
|
| 2090 |
+
},
|
| 2091 |
+
{
|
| 2092 |
+
"epoch": 2.521150592216582,
|
| 2093 |
+
"grad_norm": 0.0015464244643226266,
|
| 2094 |
+
"learning_rate": 3.1979695431472087e-06,
|
| 2095 |
+
"loss": 0.0002,
|
| 2096 |
+
"step": 2980
|
| 2097 |
+
},
|
| 2098 |
+
{
|
| 2099 |
+
"epoch": 2.529610829103215,
|
| 2100 |
+
"grad_norm": 0.0016948716947808862,
|
| 2101 |
+
"learning_rate": 3.1415679639029893e-06,
|
| 2102 |
+
"loss": 0.0002,
|
| 2103 |
+
"step": 2990
|
| 2104 |
+
},
|
| 2105 |
+
{
|
| 2106 |
+
"epoch": 2.5380710659898478,
|
| 2107 |
+
"grad_norm": 0.0015749187441542745,
|
| 2108 |
+
"learning_rate": 3.085166384658771e-06,
|
| 2109 |
+
"loss": 0.0002,
|
| 2110 |
+
"step": 3000
|
| 2111 |
+
},
|
| 2112 |
+
{
|
| 2113 |
+
"epoch": 2.5465313028764807,
|
| 2114 |
+
"grad_norm": 0.001559158437885344,
|
| 2115 |
+
"learning_rate": 3.0287648054145517e-06,
|
| 2116 |
+
"loss": 0.0002,
|
| 2117 |
+
"step": 3010
|
| 2118 |
+
},
|
| 2119 |
+
{
|
| 2120 |
+
"epoch": 2.5549915397631136,
|
| 2121 |
+
"grad_norm": 0.001560594537295401,
|
| 2122 |
+
"learning_rate": 2.9723632261703327e-06,
|
| 2123 |
+
"loss": 0.0002,
|
| 2124 |
+
"step": 3020
|
| 2125 |
+
},
|
| 2126 |
+
{
|
| 2127 |
+
"epoch": 2.563451776649746,
|
| 2128 |
+
"grad_norm": 0.0017236259300261736,
|
| 2129 |
+
"learning_rate": 2.915961646926114e-06,
|
| 2130 |
+
"loss": 0.0002,
|
| 2131 |
+
"step": 3030
|
| 2132 |
+
},
|
| 2133 |
+
{
|
| 2134 |
+
"epoch": 2.571912013536379,
|
| 2135 |
+
"grad_norm": 0.001570373773574829,
|
| 2136 |
+
"learning_rate": 2.859560067681895e-06,
|
| 2137 |
+
"loss": 0.0002,
|
| 2138 |
+
"step": 3040
|
| 2139 |
+
},
|
| 2140 |
+
{
|
| 2141 |
+
"epoch": 2.5803722504230118,
|
| 2142 |
+
"grad_norm": 0.0015271154697984457,
|
| 2143 |
+
"learning_rate": 2.8031584884376766e-06,
|
| 2144 |
+
"loss": 0.0002,
|
| 2145 |
+
"step": 3050
|
| 2146 |
+
},
|
| 2147 |
+
{
|
| 2148 |
+
"epoch": 2.5888324873096447,
|
| 2149 |
+
"grad_norm": 0.0015824285801500082,
|
| 2150 |
+
"learning_rate": 2.7467569091934576e-06,
|
| 2151 |
+
"loss": 0.0002,
|
| 2152 |
+
"step": 3060
|
| 2153 |
+
},
|
| 2154 |
+
{
|
| 2155 |
+
"epoch": 2.5972927241962775,
|
| 2156 |
+
"grad_norm": 0.0015454943059012294,
|
| 2157 |
+
"learning_rate": 2.6903553299492387e-06,
|
| 2158 |
+
"loss": 0.0002,
|
| 2159 |
+
"step": 3070
|
| 2160 |
+
},
|
| 2161 |
+
{
|
| 2162 |
+
"epoch": 2.6057529610829104,
|
| 2163 |
+
"grad_norm": 0.0014993359800428152,
|
| 2164 |
+
"learning_rate": 2.63395375070502e-06,
|
| 2165 |
+
"loss": 0.0002,
|
| 2166 |
+
"step": 3080
|
| 2167 |
+
},
|
| 2168 |
+
{
|
| 2169 |
+
"epoch": 2.6142131979695433,
|
| 2170 |
+
"grad_norm": 0.0015315411146730185,
|
| 2171 |
+
"learning_rate": 2.577552171460801e-06,
|
| 2172 |
+
"loss": 0.0002,
|
| 2173 |
+
"step": 3090
|
| 2174 |
+
},
|
| 2175 |
+
{
|
| 2176 |
+
"epoch": 2.6226734348561758,
|
| 2177 |
+
"grad_norm": 0.001639287336729467,
|
| 2178 |
+
"learning_rate": 2.5211505922165825e-06,
|
| 2179 |
+
"loss": 0.0002,
|
| 2180 |
+
"step": 3100
|
| 2181 |
+
},
|
| 2182 |
+
{
|
| 2183 |
+
"epoch": 2.6311336717428087,
|
| 2184 |
+
"grad_norm": 0.0015661792131140828,
|
| 2185 |
+
"learning_rate": 2.4647490129723635e-06,
|
| 2186 |
+
"loss": 0.0002,
|
| 2187 |
+
"step": 3110
|
| 2188 |
+
},
|
| 2189 |
+
{
|
| 2190 |
+
"epoch": 2.6395939086294415,
|
| 2191 |
+
"grad_norm": 0.0015322198159992695,
|
| 2192 |
+
"learning_rate": 2.4083474337281446e-06,
|
| 2193 |
+
"loss": 0.0002,
|
| 2194 |
+
"step": 3120
|
| 2195 |
+
},
|
| 2196 |
+
{
|
| 2197 |
+
"epoch": 2.6480541455160744,
|
| 2198 |
+
"grad_norm": 0.0015318789519369602,
|
| 2199 |
+
"learning_rate": 2.3519458544839256e-06,
|
| 2200 |
+
"loss": 0.0002,
|
| 2201 |
+
"step": 3130
|
| 2202 |
+
},
|
| 2203 |
+
{
|
| 2204 |
+
"epoch": 2.6565143824027073,
|
| 2205 |
+
"grad_norm": 0.3914497494697571,
|
| 2206 |
+
"learning_rate": 2.295544275239707e-06,
|
| 2207 |
+
"loss": 0.0002,
|
| 2208 |
+
"step": 3140
|
| 2209 |
+
},
|
| 2210 |
+
{
|
| 2211 |
+
"epoch": 2.66497461928934,
|
| 2212 |
+
"grad_norm": 0.001593979774042964,
|
| 2213 |
+
"learning_rate": 2.239142695995488e-06,
|
| 2214 |
+
"loss": 0.0002,
|
| 2215 |
+
"step": 3150
|
| 2216 |
+
},
|
| 2217 |
+
{
|
| 2218 |
+
"epoch": 2.673434856175973,
|
| 2219 |
+
"grad_norm": 0.0015341712860390544,
|
| 2220 |
+
"learning_rate": 2.182741116751269e-06,
|
| 2221 |
+
"loss": 0.0002,
|
| 2222 |
+
"step": 3160
|
| 2223 |
+
},
|
| 2224 |
+
{
|
| 2225 |
+
"epoch": 2.6818950930626055,
|
| 2226 |
+
"grad_norm": 0.001517930650152266,
|
| 2227 |
+
"learning_rate": 2.1263395375070505e-06,
|
| 2228 |
+
"loss": 0.0002,
|
| 2229 |
+
"step": 3170
|
| 2230 |
+
},
|
| 2231 |
+
{
|
| 2232 |
+
"epoch": 2.6903553299492384,
|
| 2233 |
+
"grad_norm": 0.0017692632973194122,
|
| 2234 |
+
"learning_rate": 2.0699379582628315e-06,
|
| 2235 |
+
"loss": 0.0002,
|
| 2236 |
+
"step": 3180
|
| 2237 |
+
},
|
| 2238 |
+
{
|
| 2239 |
+
"epoch": 2.6988155668358713,
|
| 2240 |
+
"grad_norm": 0.0015174917643889785,
|
| 2241 |
+
"learning_rate": 2.0135363790186125e-06,
|
| 2242 |
+
"loss": 0.0002,
|
| 2243 |
+
"step": 3190
|
| 2244 |
+
},
|
| 2245 |
+
{
|
| 2246 |
+
"epoch": 2.707275803722504,
|
| 2247 |
+
"grad_norm": 0.0015545767964795232,
|
| 2248 |
+
"learning_rate": 1.957134799774394e-06,
|
| 2249 |
+
"loss": 0.0002,
|
| 2250 |
+
"step": 3200
|
| 2251 |
+
},
|
| 2252 |
+
{
|
| 2253 |
+
"epoch": 2.715736040609137,
|
| 2254 |
+
"grad_norm": 0.0015691625885665417,
|
| 2255 |
+
"learning_rate": 1.9007332205301752e-06,
|
| 2256 |
+
"loss": 0.0002,
|
| 2257 |
+
"step": 3210
|
| 2258 |
+
},
|
| 2259 |
+
{
|
| 2260 |
+
"epoch": 2.72419627749577,
|
| 2261 |
+
"grad_norm": 0.0015581439947709441,
|
| 2262 |
+
"learning_rate": 1.844331641285956e-06,
|
| 2263 |
+
"loss": 0.0002,
|
| 2264 |
+
"step": 3220
|
| 2265 |
+
},
|
| 2266 |
+
{
|
| 2267 |
+
"epoch": 2.732656514382403,
|
| 2268 |
+
"grad_norm": 0.001601015916094184,
|
| 2269 |
+
"learning_rate": 1.7879300620417372e-06,
|
| 2270 |
+
"loss": 0.0002,
|
| 2271 |
+
"step": 3230
|
| 2272 |
+
},
|
| 2273 |
+
{
|
| 2274 |
+
"epoch": 2.7411167512690353,
|
| 2275 |
+
"grad_norm": 0.0015418503899127245,
|
| 2276 |
+
"learning_rate": 1.7315284827975184e-06,
|
| 2277 |
+
"loss": 0.0002,
|
| 2278 |
+
"step": 3240
|
| 2279 |
+
},
|
| 2280 |
+
{
|
| 2281 |
+
"epoch": 2.749576988155668,
|
| 2282 |
+
"grad_norm": 0.0014788530534133315,
|
| 2283 |
+
"learning_rate": 1.6751269035532996e-06,
|
| 2284 |
+
"loss": 0.0002,
|
| 2285 |
+
"step": 3250
|
| 2286 |
+
},
|
| 2287 |
+
{
|
| 2288 |
+
"epoch": 2.758037225042301,
|
| 2289 |
+
"grad_norm": 0.001493146293796599,
|
| 2290 |
+
"learning_rate": 1.6187253243090809e-06,
|
| 2291 |
+
"loss": 0.0002,
|
| 2292 |
+
"step": 3260
|
| 2293 |
+
},
|
| 2294 |
+
{
|
| 2295 |
+
"epoch": 2.766497461928934,
|
| 2296 |
+
"grad_norm": 0.001555291237309575,
|
| 2297 |
+
"learning_rate": 1.562323745064862e-06,
|
| 2298 |
+
"loss": 0.0002,
|
| 2299 |
+
"step": 3270
|
| 2300 |
+
},
|
| 2301 |
+
{
|
| 2302 |
+
"epoch": 2.774957698815567,
|
| 2303 |
+
"grad_norm": 0.0014956113882362843,
|
| 2304 |
+
"learning_rate": 1.5059221658206429e-06,
|
| 2305 |
+
"loss": 0.0002,
|
| 2306 |
+
"step": 3280
|
| 2307 |
+
},
|
| 2308 |
+
{
|
| 2309 |
+
"epoch": 2.7834179357021998,
|
| 2310 |
+
"grad_norm": 0.0014980457490310073,
|
| 2311 |
+
"learning_rate": 1.4495205865764241e-06,
|
| 2312 |
+
"loss": 0.0002,
|
| 2313 |
+
"step": 3290
|
| 2314 |
+
},
|
| 2315 |
+
{
|
| 2316 |
+
"epoch": 2.7918781725888326,
|
| 2317 |
+
"grad_norm": 0.0015487979399040341,
|
| 2318 |
+
"learning_rate": 1.3931190073322053e-06,
|
| 2319 |
+
"loss": 0.0002,
|
| 2320 |
+
"step": 3300
|
| 2321 |
+
},
|
| 2322 |
+
{
|
| 2323 |
+
"epoch": 2.800338409475465,
|
| 2324 |
+
"grad_norm": 0.0015123431803658605,
|
| 2325 |
+
"learning_rate": 1.3367174280879866e-06,
|
| 2326 |
+
"loss": 0.0002,
|
| 2327 |
+
"step": 3310
|
| 2328 |
+
},
|
| 2329 |
+
{
|
| 2330 |
+
"epoch": 2.808798646362098,
|
| 2331 |
+
"grad_norm": 0.0015372063498944044,
|
| 2332 |
+
"learning_rate": 1.2803158488437678e-06,
|
| 2333 |
+
"loss": 0.0002,
|
| 2334 |
+
"step": 3320
|
| 2335 |
+
},
|
| 2336 |
+
{
|
| 2337 |
+
"epoch": 2.817258883248731,
|
| 2338 |
+
"grad_norm": 0.0014844317920506,
|
| 2339 |
+
"learning_rate": 1.2239142695995488e-06,
|
| 2340 |
+
"loss": 0.0002,
|
| 2341 |
+
"step": 3330
|
| 2342 |
+
},
|
| 2343 |
+
{
|
| 2344 |
+
"epoch": 2.8257191201353637,
|
| 2345 |
+
"grad_norm": 0.0015311125898733735,
|
| 2346 |
+
"learning_rate": 1.16751269035533e-06,
|
| 2347 |
+
"loss": 0.0002,
|
| 2348 |
+
"step": 3340
|
| 2349 |
+
},
|
| 2350 |
+
{
|
| 2351 |
+
"epoch": 2.8341793570219966,
|
| 2352 |
+
"grad_norm": 0.0014878514921292663,
|
| 2353 |
+
"learning_rate": 1.111111111111111e-06,
|
| 2354 |
+
"loss": 0.0002,
|
| 2355 |
+
"step": 3350
|
| 2356 |
+
},
|
| 2357 |
+
{
|
| 2358 |
+
"epoch": 2.8426395939086295,
|
| 2359 |
+
"grad_norm": 0.0015007068868726492,
|
| 2360 |
+
"learning_rate": 1.0547095318668923e-06,
|
| 2361 |
+
"loss": 0.0002,
|
| 2362 |
+
"step": 3360
|
| 2363 |
+
},
|
| 2364 |
+
{
|
| 2365 |
+
"epoch": 2.8510998307952624,
|
| 2366 |
+
"grad_norm": 0.0014825836988165975,
|
| 2367 |
+
"learning_rate": 9.983079526226735e-07,
|
| 2368 |
+
"loss": 0.0002,
|
| 2369 |
+
"step": 3370
|
| 2370 |
+
},
|
| 2371 |
+
{
|
| 2372 |
+
"epoch": 2.859560067681895,
|
| 2373 |
+
"grad_norm": 0.0015565445646643639,
|
| 2374 |
+
"learning_rate": 9.419063733784546e-07,
|
| 2375 |
+
"loss": 0.0002,
|
| 2376 |
+
"step": 3380
|
| 2377 |
+
},
|
| 2378 |
+
{
|
| 2379 |
+
"epoch": 2.868020304568528,
|
| 2380 |
+
"grad_norm": 0.00148012675344944,
|
| 2381 |
+
"learning_rate": 8.855047941342358e-07,
|
| 2382 |
+
"loss": 0.0002,
|
| 2383 |
+
"step": 3390
|
| 2384 |
+
},
|
| 2385 |
+
{
|
| 2386 |
+
"epoch": 2.8764805414551606,
|
| 2387 |
+
"grad_norm": 0.0015161291230469942,
|
| 2388 |
+
"learning_rate": 8.291032148900171e-07,
|
| 2389 |
+
"loss": 0.0002,
|
| 2390 |
+
"step": 3400
|
| 2391 |
+
},
|
| 2392 |
+
{
|
| 2393 |
+
"epoch": 2.8849407783417935,
|
| 2394 |
+
"grad_norm": 0.0015282456297427416,
|
| 2395 |
+
"learning_rate": 7.727016356457981e-07,
|
| 2396 |
+
"loss": 0.0002,
|
| 2397 |
+
"step": 3410
|
| 2398 |
+
},
|
| 2399 |
+
{
|
| 2400 |
+
"epoch": 2.8934010152284264,
|
| 2401 |
+
"grad_norm": 0.0014796694740653038,
|
| 2402 |
+
"learning_rate": 7.163000564015793e-07,
|
| 2403 |
+
"loss": 0.0002,
|
| 2404 |
+
"step": 3420
|
| 2405 |
+
},
|
| 2406 |
+
{
|
| 2407 |
+
"epoch": 2.9018612521150593,
|
| 2408 |
+
"grad_norm": 0.0014780033379793167,
|
| 2409 |
+
"learning_rate": 6.598984771573605e-07,
|
| 2410 |
+
"loss": 0.0002,
|
| 2411 |
+
"step": 3430
|
| 2412 |
+
},
|
| 2413 |
+
{
|
| 2414 |
+
"epoch": 2.910321489001692,
|
| 2415 |
+
"grad_norm": 0.0014671003445982933,
|
| 2416 |
+
"learning_rate": 6.034968979131416e-07,
|
| 2417 |
+
"loss": 0.0002,
|
| 2418 |
+
"step": 3440
|
| 2419 |
+
},
|
| 2420 |
+
{
|
| 2421 |
+
"epoch": 2.9187817258883246,
|
| 2422 |
+
"grad_norm": 0.0016121144872158766,
|
| 2423 |
+
"learning_rate": 5.470953186689228e-07,
|
| 2424 |
+
"loss": 0.0002,
|
| 2425 |
+
"step": 3450
|
| 2426 |
+
},
|
| 2427 |
+
{
|
| 2428 |
+
"epoch": 2.927241962774958,
|
| 2429 |
+
"grad_norm": 0.0015159000176936388,
|
| 2430 |
+
"learning_rate": 4.906937394247039e-07,
|
| 2431 |
+
"loss": 0.0002,
|
| 2432 |
+
"step": 3460
|
| 2433 |
+
},
|
| 2434 |
+
{
|
| 2435 |
+
"epoch": 2.9357021996615904,
|
| 2436 |
+
"grad_norm": 0.0015044745523482561,
|
| 2437 |
+
"learning_rate": 4.342921601804851e-07,
|
| 2438 |
+
"loss": 0.0005,
|
| 2439 |
+
"step": 3470
|
| 2440 |
+
},
|
| 2441 |
+
{
|
| 2442 |
+
"epoch": 2.9441624365482233,
|
| 2443 |
+
"grad_norm": 0.029072463512420654,
|
| 2444 |
+
"learning_rate": 3.778905809362662e-07,
|
| 2445 |
+
"loss": 0.0002,
|
| 2446 |
+
"step": 3480
|
| 2447 |
+
},
|
| 2448 |
+
{
|
| 2449 |
+
"epoch": 2.952622673434856,
|
| 2450 |
+
"grad_norm": 0.0014558079419657588,
|
| 2451 |
+
"learning_rate": 3.214890016920474e-07,
|
| 2452 |
+
"loss": 0.0002,
|
| 2453 |
+
"step": 3490
|
| 2454 |
+
},
|
| 2455 |
+
{
|
| 2456 |
+
"epoch": 2.961082910321489,
|
| 2457 |
+
"grad_norm": 0.0015029908390715718,
|
| 2458 |
+
"learning_rate": 2.6508742244782856e-07,
|
| 2459 |
+
"loss": 0.0002,
|
| 2460 |
+
"step": 3500
|
| 2461 |
+
},
|
| 2462 |
+
{
|
| 2463 |
+
"epoch": 2.969543147208122,
|
| 2464 |
+
"grad_norm": 0.001473087933845818,
|
| 2465 |
+
"learning_rate": 2.086858432036097e-07,
|
| 2466 |
+
"loss": 0.0002,
|
| 2467 |
+
"step": 3510
|
| 2468 |
+
},
|
| 2469 |
+
{
|
| 2470 |
+
"epoch": 2.9780033840947544,
|
| 2471 |
+
"grad_norm": 0.001540781813673675,
|
| 2472 |
+
"learning_rate": 1.5228426395939088e-07,
|
| 2473 |
+
"loss": 0.0002,
|
| 2474 |
+
"step": 3520
|
| 2475 |
+
},
|
| 2476 |
+
{
|
| 2477 |
+
"epoch": 2.9864636209813877,
|
| 2478 |
+
"grad_norm": 0.0014602647861465812,
|
| 2479 |
+
"learning_rate": 9.588268471517204e-08,
|
| 2480 |
+
"loss": 0.0002,
|
| 2481 |
+
"step": 3530
|
| 2482 |
+
},
|
| 2483 |
+
{
|
| 2484 |
+
"epoch": 2.99492385786802,
|
| 2485 |
+
"grad_norm": 0.0015418573748320341,
|
| 2486 |
+
"learning_rate": 3.9481105470953196e-08,
|
| 2487 |
+
"loss": 0.0002,
|
| 2488 |
+
"step": 3540
|
| 2489 |
+
}
|
| 2490 |
+
],
|
| 2491 |
+
"logging_steps": 10,
|
| 2492 |
+
"max_steps": 3546,
|
| 2493 |
+
"num_input_tokens_seen": 0,
|
| 2494 |
+
"num_train_epochs": 3,
|
| 2495 |
+
"save_steps": 500,
|
| 2496 |
+
"stateful_callbacks": {
|
| 2497 |
+
"TrainerControl": {
|
| 2498 |
+
"args": {
|
| 2499 |
+
"should_epoch_stop": false,
|
| 2500 |
+
"should_evaluate": false,
|
| 2501 |
+
"should_log": false,
|
| 2502 |
+
"should_save": true,
|
| 2503 |
+
"should_training_stop": true
|
| 2504 |
+
},
|
| 2505 |
+
"attributes": {}
|
| 2506 |
+
}
|
| 2507 |
+
},
|
| 2508 |
+
"total_flos": 1058452130079540.0,
|
| 2509 |
+
"train_batch_size": 8,
|
| 2510 |
+
"trial_name": null,
|
| 2511 |
+
"trial_params": null
|
| 2512 |
+
}
|