Upload folder using huggingface_hub
Browse files- optimizer.pt +3 -0
- pytorch_model.bin +2 -2
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- trainer_state.json +1433 -0
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9f31cc85c5861b4342c1ec63c798cc321445f9274bf7bc8265474da8d1ca55fe
|
| 3 |
+
size 2460081018
|
pytorch_model.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7d9368701771c80702b6761ea87bef83ad34648eb1c7d47205d35b16e2cc2a3c
|
| 3 |
+
size 1244204446
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0a78221583471cad386fd48e42b8f2c9cf2296c8f66a01b40f0d76a42c7fcea5
|
| 3 |
+
size 14244
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0280f775a6a8bd177e6c3996391bae071c99817f5b9d8602ebe3697eddfe7d37
|
| 3 |
+
size 1064
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1433 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.1999980000199998,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 20000,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.000999990000099999,
|
| 13 |
+
"grad_norm": 5341.90380859375,
|
| 14 |
+
"learning_rate": 2.0000000000000003e-06,
|
| 15 |
+
"loss": 594.5605,
|
| 16 |
+
"step": 100
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.001999980000199998,
|
| 20 |
+
"grad_norm": 789.9061889648438,
|
| 21 |
+
"learning_rate": 4.000000000000001e-06,
|
| 22 |
+
"loss": 127.941,
|
| 23 |
+
"step": 200
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.002999970000299997,
|
| 27 |
+
"grad_norm": 306.2316589355469,
|
| 28 |
+
"learning_rate": 6e-06,
|
| 29 |
+
"loss": 119.2364,
|
| 30 |
+
"step": 300
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.003999960000399996,
|
| 34 |
+
"grad_norm": 751.376220703125,
|
| 35 |
+
"learning_rate": 8.000000000000001e-06,
|
| 36 |
+
"loss": 118.0761,
|
| 37 |
+
"step": 400
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.004999950000499995,
|
| 41 |
+
"grad_norm": 335.5125732421875,
|
| 42 |
+
"learning_rate": 1e-05,
|
| 43 |
+
"loss": 109.0291,
|
| 44 |
+
"step": 500
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.005999940000599994,
|
| 48 |
+
"grad_norm": 510.80975341796875,
|
| 49 |
+
"learning_rate": 1.2e-05,
|
| 50 |
+
"loss": 109.4396,
|
| 51 |
+
"step": 600
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.006999930000699993,
|
| 55 |
+
"grad_norm": 314.0848083496094,
|
| 56 |
+
"learning_rate": 1.4e-05,
|
| 57 |
+
"loss": 84.1079,
|
| 58 |
+
"step": 700
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.007999920000799993,
|
| 62 |
+
"grad_norm": 935.3131713867188,
|
| 63 |
+
"learning_rate": 1.6000000000000003e-05,
|
| 64 |
+
"loss": 82.2216,
|
| 65 |
+
"step": 800
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.008999910000899992,
|
| 69 |
+
"grad_norm": 1358.540771484375,
|
| 70 |
+
"learning_rate": 1.8e-05,
|
| 71 |
+
"loss": 80.6757,
|
| 72 |
+
"step": 900
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.00999990000099999,
|
| 76 |
+
"grad_norm": 317.9821472167969,
|
| 77 |
+
"learning_rate": 2e-05,
|
| 78 |
+
"loss": 62.5531,
|
| 79 |
+
"step": 1000
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.010999890001099988,
|
| 83 |
+
"grad_norm": 649.0853271484375,
|
| 84 |
+
"learning_rate": 2.2000000000000003e-05,
|
| 85 |
+
"loss": 56.7378,
|
| 86 |
+
"step": 1100
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.011999880001199987,
|
| 90 |
+
"grad_norm": 530.7359008789062,
|
| 91 |
+
"learning_rate": 2.4e-05,
|
| 92 |
+
"loss": 56.3481,
|
| 93 |
+
"step": 1200
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.012999870001299986,
|
| 97 |
+
"grad_norm": 241.44517517089844,
|
| 98 |
+
"learning_rate": 2.6000000000000002e-05,
|
| 99 |
+
"loss": 53.1624,
|
| 100 |
+
"step": 1300
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.013999860001399985,
|
| 104 |
+
"grad_norm": 1705.9263916015625,
|
| 105 |
+
"learning_rate": 2.8e-05,
|
| 106 |
+
"loss": 57.0686,
|
| 107 |
+
"step": 1400
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.014999850001499985,
|
| 111 |
+
"grad_norm": 1379.0430908203125,
|
| 112 |
+
"learning_rate": 3.0000000000000004e-05,
|
| 113 |
+
"loss": 47.7146,
|
| 114 |
+
"step": 1500
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.015999840001599985,
|
| 118 |
+
"grad_norm": 253.24891662597656,
|
| 119 |
+
"learning_rate": 3.2000000000000005e-05,
|
| 120 |
+
"loss": 42.4126,
|
| 121 |
+
"step": 1600
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.016999830001699984,
|
| 125 |
+
"grad_norm": 258.5624694824219,
|
| 126 |
+
"learning_rate": 3.4e-05,
|
| 127 |
+
"loss": 48.3719,
|
| 128 |
+
"step": 1700
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.017999820001799983,
|
| 132 |
+
"grad_norm": 273.0712585449219,
|
| 133 |
+
"learning_rate": 3.6e-05,
|
| 134 |
+
"loss": 38.7285,
|
| 135 |
+
"step": 1800
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.018999810001899983,
|
| 139 |
+
"grad_norm": 820.5573120117188,
|
| 140 |
+
"learning_rate": 3.8e-05,
|
| 141 |
+
"loss": 44.3776,
|
| 142 |
+
"step": 1900
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.01999980000199998,
|
| 146 |
+
"grad_norm": 538.52587890625,
|
| 147 |
+
"learning_rate": 4e-05,
|
| 148 |
+
"loss": 33.6354,
|
| 149 |
+
"step": 2000
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.02099979000209998,
|
| 153 |
+
"grad_norm": 987.8306884765625,
|
| 154 |
+
"learning_rate": 3.999997482501191e-05,
|
| 155 |
+
"loss": 39.2998,
|
| 156 |
+
"step": 2100
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.021999780002199976,
|
| 160 |
+
"grad_norm": 963.3218383789062,
|
| 161 |
+
"learning_rate": 3.9999899300111024e-05,
|
| 162 |
+
"loss": 30.5145,
|
| 163 |
+
"step": 2200
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.022999770002299975,
|
| 167 |
+
"grad_norm": 384.54278564453125,
|
| 168 |
+
"learning_rate": 3.999977342548747e-05,
|
| 169 |
+
"loss": 26.2319,
|
| 170 |
+
"step": 2300
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.023999760002399975,
|
| 174 |
+
"grad_norm": 362.4068908691406,
|
| 175 |
+
"learning_rate": 3.9999597201458134e-05,
|
| 176 |
+
"loss": 21.7486,
|
| 177 |
+
"step": 2400
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.024999750002499974,
|
| 181 |
+
"grad_norm": 246.06344604492188,
|
| 182 |
+
"learning_rate": 3.9999370628466666e-05,
|
| 183 |
+
"loss": 18.8825,
|
| 184 |
+
"step": 2500
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.025999740002599973,
|
| 188 |
+
"grad_norm": 558.910888671875,
|
| 189 |
+
"learning_rate": 3.9999093707083455e-05,
|
| 190 |
+
"loss": 18.5258,
|
| 191 |
+
"step": 2600
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.026999730002699972,
|
| 195 |
+
"grad_norm": 4613.8505859375,
|
| 196 |
+
"learning_rate": 3.999876643800567e-05,
|
| 197 |
+
"loss": 20.1839,
|
| 198 |
+
"step": 2700
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.02799972000279997,
|
| 202 |
+
"grad_norm": 246.72665405273438,
|
| 203 |
+
"learning_rate": 3.999838882205719e-05,
|
| 204 |
+
"loss": 17.0552,
|
| 205 |
+
"step": 2800
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.02899971000289997,
|
| 209 |
+
"grad_norm": 85.92310333251953,
|
| 210 |
+
"learning_rate": 3.9997960860188666e-05,
|
| 211 |
+
"loss": 16.4099,
|
| 212 |
+
"step": 2900
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.02999970000299997,
|
| 216 |
+
"grad_norm": 607.7698364257812,
|
| 217 |
+
"learning_rate": 3.9997482553477506e-05,
|
| 218 |
+
"loss": 15.0478,
|
| 219 |
+
"step": 3000
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.030999690003099968,
|
| 223 |
+
"grad_norm": 348.2640075683594,
|
| 224 |
+
"learning_rate": 3.999695390312783e-05,
|
| 225 |
+
"loss": 13.6557,
|
| 226 |
+
"step": 3100
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.03199968000319997,
|
| 230 |
+
"grad_norm": 553.7791137695312,
|
| 231 |
+
"learning_rate": 3.999637491047052e-05,
|
| 232 |
+
"loss": 16.3278,
|
| 233 |
+
"step": 3200
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.032999670003299966,
|
| 237 |
+
"grad_norm": 851.149169921875,
|
| 238 |
+
"learning_rate": 3.999574557696319e-05,
|
| 239 |
+
"loss": 15.7218,
|
| 240 |
+
"step": 3300
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.03399966000339997,
|
| 244 |
+
"grad_norm": 131.01693725585938,
|
| 245 |
+
"learning_rate": 3.9995065904190185e-05,
|
| 246 |
+
"loss": 11.3467,
|
| 247 |
+
"step": 3400
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.034999650003499964,
|
| 251 |
+
"grad_norm": 41.529632568359375,
|
| 252 |
+
"learning_rate": 3.999433589386259e-05,
|
| 253 |
+
"loss": 12.6374,
|
| 254 |
+
"step": 3500
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.03599964000359997,
|
| 258 |
+
"grad_norm": 205.26910400390625,
|
| 259 |
+
"learning_rate": 3.9993555547818186e-05,
|
| 260 |
+
"loss": 13.2414,
|
| 261 |
+
"step": 3600
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.03699963000369996,
|
| 265 |
+
"grad_norm": 89.92840576171875,
|
| 266 |
+
"learning_rate": 3.999272486802151e-05,
|
| 267 |
+
"loss": 10.1512,
|
| 268 |
+
"step": 3700
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.037999620003799965,
|
| 272 |
+
"grad_norm": 410.4055480957031,
|
| 273 |
+
"learning_rate": 3.9991843856563786e-05,
|
| 274 |
+
"loss": 9.8745,
|
| 275 |
+
"step": 3800
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.03899961000389996,
|
| 279 |
+
"grad_norm": 296.5409240722656,
|
| 280 |
+
"learning_rate": 3.999091251566297e-05,
|
| 281 |
+
"loss": 12.645,
|
| 282 |
+
"step": 3900
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.03999960000399996,
|
| 286 |
+
"grad_norm": 49.79493713378906,
|
| 287 |
+
"learning_rate": 3.9989930847663706e-05,
|
| 288 |
+
"loss": 12.447,
|
| 289 |
+
"step": 4000
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.04099959000409996,
|
| 293 |
+
"grad_norm": 477.4189147949219,
|
| 294 |
+
"learning_rate": 3.998889885503734e-05,
|
| 295 |
+
"loss": 11.1971,
|
| 296 |
+
"step": 4100
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.04199958000419996,
|
| 300 |
+
"grad_norm": 236.76950073242188,
|
| 301 |
+
"learning_rate": 3.998781654038192e-05,
|
| 302 |
+
"loss": 11.6718,
|
| 303 |
+
"step": 4200
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.04299957000429996,
|
| 307 |
+
"grad_norm": 484.04681396484375,
|
| 308 |
+
"learning_rate": 3.998668390642216e-05,
|
| 309 |
+
"loss": 9.9448,
|
| 310 |
+
"step": 4300
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.04399956000439995,
|
| 314 |
+
"grad_norm": 313.50457763671875,
|
| 315 |
+
"learning_rate": 3.998550095600948e-05,
|
| 316 |
+
"loss": 10.5262,
|
| 317 |
+
"step": 4400
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.044999550004499955,
|
| 321 |
+
"grad_norm": 82.21493530273438,
|
| 322 |
+
"learning_rate": 3.998426769212194e-05,
|
| 323 |
+
"loss": 13.1368,
|
| 324 |
+
"step": 4500
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.04599954000459995,
|
| 328 |
+
"grad_norm": 151.22616577148438,
|
| 329 |
+
"learning_rate": 3.9982984117864285e-05,
|
| 330 |
+
"loss": 9.8138,
|
| 331 |
+
"step": 4600
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.04699953000469995,
|
| 335 |
+
"grad_norm": 274.2380676269531,
|
| 336 |
+
"learning_rate": 3.9981650236467916e-05,
|
| 337 |
+
"loss": 10.2387,
|
| 338 |
+
"step": 4700
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.04799952000479995,
|
| 342 |
+
"grad_norm": 77.92356872558594,
|
| 343 |
+
"learning_rate": 3.998026605129088e-05,
|
| 344 |
+
"loss": 9.5225,
|
| 345 |
+
"step": 4800
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.04899951000489995,
|
| 349 |
+
"grad_norm": 202.85816955566406,
|
| 350 |
+
"learning_rate": 3.997883156581786e-05,
|
| 351 |
+
"loss": 9.8305,
|
| 352 |
+
"step": 4900
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.04999950000499995,
|
| 356 |
+
"grad_norm": 121.39033508300781,
|
| 357 |
+
"learning_rate": 3.9977346783660165e-05,
|
| 358 |
+
"loss": 12.1433,
|
| 359 |
+
"step": 5000
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.05099949000509995,
|
| 363 |
+
"grad_norm": 703.3681030273438,
|
| 364 |
+
"learning_rate": 3.997581170855573e-05,
|
| 365 |
+
"loss": 9.3006,
|
| 366 |
+
"step": 5100
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.051999480005199945,
|
| 370 |
+
"grad_norm": 905.860107421875,
|
| 371 |
+
"learning_rate": 3.9974226344369124e-05,
|
| 372 |
+
"loss": 9.2302,
|
| 373 |
+
"step": 5200
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.05299947000529995,
|
| 377 |
+
"grad_norm": 108.33662414550781,
|
| 378 |
+
"learning_rate": 3.9972590695091476e-05,
|
| 379 |
+
"loss": 9.0692,
|
| 380 |
+
"step": 5300
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.053999460005399944,
|
| 384 |
+
"grad_norm": 278.9888916015625,
|
| 385 |
+
"learning_rate": 3.9970904764840554e-05,
|
| 386 |
+
"loss": 9.0078,
|
| 387 |
+
"step": 5400
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.054999450005499946,
|
| 391 |
+
"grad_norm": 339.5447692871094,
|
| 392 |
+
"learning_rate": 3.9969168557860665e-05,
|
| 393 |
+
"loss": 8.5263,
|
| 394 |
+
"step": 5500
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.05599944000559994,
|
| 398 |
+
"grad_norm": 479.4326477050781,
|
| 399 |
+
"learning_rate": 3.9967382078522716e-05,
|
| 400 |
+
"loss": 8.6827,
|
| 401 |
+
"step": 5600
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.056999430005699944,
|
| 405 |
+
"grad_norm": 205.8404083251953,
|
| 406 |
+
"learning_rate": 3.9965545331324166e-05,
|
| 407 |
+
"loss": 8.0084,
|
| 408 |
+
"step": 5700
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.05799942000579994,
|
| 412 |
+
"grad_norm": 318.3268127441406,
|
| 413 |
+
"learning_rate": 3.996365832088903e-05,
|
| 414 |
+
"loss": 7.8264,
|
| 415 |
+
"step": 5800
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.05899941000589994,
|
| 419 |
+
"grad_norm": 545.3519287109375,
|
| 420 |
+
"learning_rate": 3.996172105196785e-05,
|
| 421 |
+
"loss": 8.5439,
|
| 422 |
+
"step": 5900
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.05999940000599994,
|
| 426 |
+
"grad_norm": 536.3674926757812,
|
| 427 |
+
"learning_rate": 3.995973352943769e-05,
|
| 428 |
+
"loss": 7.2916,
|
| 429 |
+
"step": 6000
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.06099939000609994,
|
| 433 |
+
"grad_norm": 29.15781593322754,
|
| 434 |
+
"learning_rate": 3.995769575830215e-05,
|
| 435 |
+
"loss": 8.3879,
|
| 436 |
+
"step": 6100
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.061999380006199936,
|
| 440 |
+
"grad_norm": 573.3207397460938,
|
| 441 |
+
"learning_rate": 3.99556077436913e-05,
|
| 442 |
+
"loss": 6.6733,
|
| 443 |
+
"step": 6200
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.06299937000629993,
|
| 447 |
+
"grad_norm": 206.36526489257812,
|
| 448 |
+
"learning_rate": 3.995346949086174e-05,
|
| 449 |
+
"loss": 5.7694,
|
| 450 |
+
"step": 6300
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.06399936000639994,
|
| 454 |
+
"grad_norm": 16834.9296875,
|
| 455 |
+
"learning_rate": 3.9951281005196486e-05,
|
| 456 |
+
"loss": 6.1864,
|
| 457 |
+
"step": 6400
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.06499935000649994,
|
| 461 |
+
"grad_norm": 401.1425476074219,
|
| 462 |
+
"learning_rate": 3.994904229220507e-05,
|
| 463 |
+
"loss": 6.1338,
|
| 464 |
+
"step": 6500
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.06599934000659993,
|
| 468 |
+
"grad_norm": 277.509765625,
|
| 469 |
+
"learning_rate": 3.994675335752345e-05,
|
| 470 |
+
"loss": 4.8342,
|
| 471 |
+
"step": 6600
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.06699933000669993,
|
| 475 |
+
"grad_norm": 544.186767578125,
|
| 476 |
+
"learning_rate": 3.9944414206914e-05,
|
| 477 |
+
"loss": 8.32,
|
| 478 |
+
"step": 6700
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.06799932000679994,
|
| 482 |
+
"grad_norm": 174.74789428710938,
|
| 483 |
+
"learning_rate": 3.994202484626555e-05,
|
| 484 |
+
"loss": 10.03,
|
| 485 |
+
"step": 6800
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.06899931000689993,
|
| 489 |
+
"grad_norm": 562.16357421875,
|
| 490 |
+
"learning_rate": 3.99395852815933e-05,
|
| 491 |
+
"loss": 7.7643,
|
| 492 |
+
"step": 6900
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.06999930000699993,
|
| 496 |
+
"grad_norm": 219.15576171875,
|
| 497 |
+
"learning_rate": 3.993709551903885e-05,
|
| 498 |
+
"loss": 8.8717,
|
| 499 |
+
"step": 7000
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.07099929000709992,
|
| 503 |
+
"grad_norm": 159.7786102294922,
|
| 504 |
+
"learning_rate": 3.993455556487018e-05,
|
| 505 |
+
"loss": 7.3802,
|
| 506 |
+
"step": 7100
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.07199928000719993,
|
| 510 |
+
"grad_norm": 224.97128295898438,
|
| 511 |
+
"learning_rate": 3.993196542548162e-05,
|
| 512 |
+
"loss": 9.665,
|
| 513 |
+
"step": 7200
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.07299927000729993,
|
| 517 |
+
"grad_norm": 691.8470458984375,
|
| 518 |
+
"learning_rate": 3.992932510739383e-05,
|
| 519 |
+
"loss": 5.8392,
|
| 520 |
+
"step": 7300
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.07399926000739993,
|
| 524 |
+
"grad_norm": 196.7902069091797,
|
| 525 |
+
"learning_rate": 3.992663461725383e-05,
|
| 526 |
+
"loss": 5.8781,
|
| 527 |
+
"step": 7400
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.07499925000749992,
|
| 531 |
+
"grad_norm": 450.8050231933594,
|
| 532 |
+
"learning_rate": 3.9923893961834914e-05,
|
| 533 |
+
"loss": 8.0027,
|
| 534 |
+
"step": 7500
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.07599924000759993,
|
| 538 |
+
"grad_norm": 182.17953491210938,
|
| 539 |
+
"learning_rate": 3.992110314803668e-05,
|
| 540 |
+
"loss": 4.6279,
|
| 541 |
+
"step": 7600
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.07699923000769993,
|
| 545 |
+
"grad_norm": 176.743896484375,
|
| 546 |
+
"learning_rate": 3.9918262182884994e-05,
|
| 547 |
+
"loss": 8.5188,
|
| 548 |
+
"step": 7700
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.07799922000779992,
|
| 552 |
+
"grad_norm": 382.1143798828125,
|
| 553 |
+
"learning_rate": 3.9915371073531995e-05,
|
| 554 |
+
"loss": 6.4761,
|
| 555 |
+
"step": 7800
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.07899921000789992,
|
| 559 |
+
"grad_norm": 513.8743286132812,
|
| 560 |
+
"learning_rate": 3.991242982725603e-05,
|
| 561 |
+
"loss": 5.1712,
|
| 562 |
+
"step": 7900
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.07999920000799993,
|
| 566 |
+
"grad_norm": 197.9111785888672,
|
| 567 |
+
"learning_rate": 3.9909438451461695e-05,
|
| 568 |
+
"loss": 7.5148,
|
| 569 |
+
"step": 8000
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.08099919000809992,
|
| 573 |
+
"grad_norm": 165.2156982421875,
|
| 574 |
+
"learning_rate": 3.990639695367977e-05,
|
| 575 |
+
"loss": 5.1054,
|
| 576 |
+
"step": 8100
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.08199918000819992,
|
| 580 |
+
"grad_norm": 162.3336944580078,
|
| 581 |
+
"learning_rate": 3.990330534156723e-05,
|
| 582 |
+
"loss": 5.3642,
|
| 583 |
+
"step": 8200
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.08299917000829991,
|
| 587 |
+
"grad_norm": 149.72377014160156,
|
| 588 |
+
"learning_rate": 3.9900163622907196e-05,
|
| 589 |
+
"loss": 7.4674,
|
| 590 |
+
"step": 8300
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.08399916000839992,
|
| 594 |
+
"grad_norm": 110.8755874633789,
|
| 595 |
+
"learning_rate": 3.9896971805608945e-05,
|
| 596 |
+
"loss": 6.2615,
|
| 597 |
+
"step": 8400
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.08499915000849992,
|
| 601 |
+
"grad_norm": 97.61107635498047,
|
| 602 |
+
"learning_rate": 3.989372989770787e-05,
|
| 603 |
+
"loss": 6.0901,
|
| 604 |
+
"step": 8500
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.08599914000859991,
|
| 608 |
+
"grad_norm": 47.84280776977539,
|
| 609 |
+
"learning_rate": 3.989043790736547e-05,
|
| 610 |
+
"loss": 6.6694,
|
| 611 |
+
"step": 8600
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.08699913000869991,
|
| 615 |
+
"grad_norm": 418.85052490234375,
|
| 616 |
+
"learning_rate": 3.988709584286933e-05,
|
| 617 |
+
"loss": 6.1093,
|
| 618 |
+
"step": 8700
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.0879991200087999,
|
| 622 |
+
"grad_norm": 402.12933349609375,
|
| 623 |
+
"learning_rate": 3.98837037126331e-05,
|
| 624 |
+
"loss": 5.6737,
|
| 625 |
+
"step": 8800
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.08899911000889991,
|
| 629 |
+
"grad_norm": 136.71066284179688,
|
| 630 |
+
"learning_rate": 3.988026152519645e-05,
|
| 631 |
+
"loss": 6.0548,
|
| 632 |
+
"step": 8900
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.08999910000899991,
|
| 636 |
+
"grad_norm": 319.9411926269531,
|
| 637 |
+
"learning_rate": 3.9876769289225084e-05,
|
| 638 |
+
"loss": 6.1571,
|
| 639 |
+
"step": 9000
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.0909990900090999,
|
| 643 |
+
"grad_norm": 216.7753143310547,
|
| 644 |
+
"learning_rate": 3.9873227013510714e-05,
|
| 645 |
+
"loss": 4.6247,
|
| 646 |
+
"step": 9100
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.0919990800091999,
|
| 650 |
+
"grad_norm": 554.5048828125,
|
| 651 |
+
"learning_rate": 3.9869634706971e-05,
|
| 652 |
+
"loss": 6.5048,
|
| 653 |
+
"step": 9200
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.09299907000929991,
|
| 657 |
+
"grad_norm": 131.7164764404297,
|
| 658 |
+
"learning_rate": 3.986599237864959e-05,
|
| 659 |
+
"loss": 6.2413,
|
| 660 |
+
"step": 9300
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.0939990600093999,
|
| 664 |
+
"grad_norm": 102.79979705810547,
|
| 665 |
+
"learning_rate": 3.9862300037716025e-05,
|
| 666 |
+
"loss": 5.7489,
|
| 667 |
+
"step": 9400
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.0949990500094999,
|
| 671 |
+
"grad_norm": 37.355709075927734,
|
| 672 |
+
"learning_rate": 3.9858557693465766e-05,
|
| 673 |
+
"loss": 5.7535,
|
| 674 |
+
"step": 9500
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.0959990400095999,
|
| 678 |
+
"grad_norm": 137.89903259277344,
|
| 679 |
+
"learning_rate": 3.985476535532018e-05,
|
| 680 |
+
"loss": 4.4851,
|
| 681 |
+
"step": 9600
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.09699903000969991,
|
| 685 |
+
"grad_norm": 174.57391357421875,
|
| 686 |
+
"learning_rate": 3.985092303282645e-05,
|
| 687 |
+
"loss": 9.4012,
|
| 688 |
+
"step": 9700
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.0979990200097999,
|
| 692 |
+
"grad_norm": 99.09408569335938,
|
| 693 |
+
"learning_rate": 3.9847030735657624e-05,
|
| 694 |
+
"loss": 7.2147,
|
| 695 |
+
"step": 9800
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.0989990100098999,
|
| 699 |
+
"grad_norm": 265.9083557128906,
|
| 700 |
+
"learning_rate": 3.984308847361257e-05,
|
| 701 |
+
"loss": 7.8453,
|
| 702 |
+
"step": 9900
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.0999990000099999,
|
| 706 |
+
"grad_norm": 240.05801391601562,
|
| 707 |
+
"learning_rate": 3.983909625661591e-05,
|
| 708 |
+
"loss": 6.4133,
|
| 709 |
+
"step": 10000
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.1009989900100999,
|
| 713 |
+
"grad_norm": 245.1486053466797,
|
| 714 |
+
"learning_rate": 3.983505409471806e-05,
|
| 715 |
+
"loss": 5.5553,
|
| 716 |
+
"step": 10100
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.1019989800101999,
|
| 720 |
+
"grad_norm": 654.8526000976562,
|
| 721 |
+
"learning_rate": 3.9830961998095146e-05,
|
| 722 |
+
"loss": 4.5454,
|
| 723 |
+
"step": 10200
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.1029989700102999,
|
| 727 |
+
"grad_norm": 49.497371673583984,
|
| 728 |
+
"learning_rate": 3.982681997704902e-05,
|
| 729 |
+
"loss": 4.0421,
|
| 730 |
+
"step": 10300
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.10399896001039989,
|
| 734 |
+
"grad_norm": 119.20033264160156,
|
| 735 |
+
"learning_rate": 3.982262804200723e-05,
|
| 736 |
+
"loss": 4.6029,
|
| 737 |
+
"step": 10400
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.1049989500104999,
|
| 741 |
+
"grad_norm": 595.1155395507812,
|
| 742 |
+
"learning_rate": 3.981838620352294e-05,
|
| 743 |
+
"loss": 3.8456,
|
| 744 |
+
"step": 10500
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.1059989400105999,
|
| 748 |
+
"grad_norm": 29.171674728393555,
|
| 749 |
+
"learning_rate": 3.9814094472275e-05,
|
| 750 |
+
"loss": 4.0763,
|
| 751 |
+
"step": 10600
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.10699893001069989,
|
| 755 |
+
"grad_norm": 359.91021728515625,
|
| 756 |
+
"learning_rate": 3.9809752859067823e-05,
|
| 757 |
+
"loss": 3.6656,
|
| 758 |
+
"step": 10700
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.10799892001079989,
|
| 762 |
+
"grad_norm": 137.34120178222656,
|
| 763 |
+
"learning_rate": 3.980536137483141e-05,
|
| 764 |
+
"loss": 5.2409,
|
| 765 |
+
"step": 10800
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.1089989100108999,
|
| 769 |
+
"grad_norm": 359.36151123046875,
|
| 770 |
+
"learning_rate": 3.9800920030621334e-05,
|
| 771 |
+
"loss": 7.5297,
|
| 772 |
+
"step": 10900
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.10999890001099989,
|
| 776 |
+
"grad_norm": 609.0127563476562,
|
| 777 |
+
"learning_rate": 3.979642883761866e-05,
|
| 778 |
+
"loss": 5.2982,
|
| 779 |
+
"step": 11000
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.11099889001109989,
|
| 783 |
+
"grad_norm": 1026.407470703125,
|
| 784 |
+
"learning_rate": 3.979188780712996e-05,
|
| 785 |
+
"loss": 5.7533,
|
| 786 |
+
"step": 11100
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.11199888001119988,
|
| 790 |
+
"grad_norm": 1161.3192138671875,
|
| 791 |
+
"learning_rate": 3.978729695058729e-05,
|
| 792 |
+
"loss": 3.8523,
|
| 793 |
+
"step": 11200
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.11299887001129989,
|
| 797 |
+
"grad_norm": 207.47286987304688,
|
| 798 |
+
"learning_rate": 3.9782656279548114e-05,
|
| 799 |
+
"loss": 5.7909,
|
| 800 |
+
"step": 11300
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.11399886001139989,
|
| 804 |
+
"grad_norm": 180.18321228027344,
|
| 805 |
+
"learning_rate": 3.9777965805695315e-05,
|
| 806 |
+
"loss": 4.8786,
|
| 807 |
+
"step": 11400
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.11499885001149988,
|
| 811 |
+
"grad_norm": 154.166259765625,
|
| 812 |
+
"learning_rate": 3.977322554083716e-05,
|
| 813 |
+
"loss": 5.9407,
|
| 814 |
+
"step": 11500
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.11599884001159988,
|
| 818 |
+
"grad_norm": 175.445556640625,
|
| 819 |
+
"learning_rate": 3.976843549690725e-05,
|
| 820 |
+
"loss": 3.8969,
|
| 821 |
+
"step": 11600
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.11699883001169989,
|
| 825 |
+
"grad_norm": 204.4251251220703,
|
| 826 |
+
"learning_rate": 3.976359568596453e-05,
|
| 827 |
+
"loss": 4.1672,
|
| 828 |
+
"step": 11700
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.11799882001179988,
|
| 832 |
+
"grad_norm": 330.901123046875,
|
| 833 |
+
"learning_rate": 3.97587061201932e-05,
|
| 834 |
+
"loss": 4.1925,
|
| 835 |
+
"step": 11800
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.11899881001189988,
|
| 839 |
+
"grad_norm": 563.4291381835938,
|
| 840 |
+
"learning_rate": 3.9753766811902756e-05,
|
| 841 |
+
"loss": 3.4035,
|
| 842 |
+
"step": 11900
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.11999880001199988,
|
| 846 |
+
"grad_norm": 57.178955078125,
|
| 847 |
+
"learning_rate": 3.974877777352789e-05,
|
| 848 |
+
"loss": 3.4773,
|
| 849 |
+
"step": 12000
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.12099879001209989,
|
| 853 |
+
"grad_norm": 94.94606018066406,
|
| 854 |
+
"learning_rate": 3.97437390176285e-05,
|
| 855 |
+
"loss": 3.5796,
|
| 856 |
+
"step": 12100
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.12199878001219988,
|
| 860 |
+
"grad_norm": 801.7224731445312,
|
| 861 |
+
"learning_rate": 3.973865055688965e-05,
|
| 862 |
+
"loss": 3.4456,
|
| 863 |
+
"step": 12200
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.12299877001229988,
|
| 867 |
+
"grad_norm": 211.559814453125,
|
| 868 |
+
"learning_rate": 3.973351240412153e-05,
|
| 869 |
+
"loss": 5.9604,
|
| 870 |
+
"step": 12300
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.12399876001239987,
|
| 874 |
+
"grad_norm": 47.33822250366211,
|
| 875 |
+
"learning_rate": 3.972832457225944e-05,
|
| 876 |
+
"loss": 1.6479,
|
| 877 |
+
"step": 12400
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.12499875001249988,
|
| 881 |
+
"grad_norm": 517.2850341796875,
|
| 882 |
+
"learning_rate": 3.972308707436374e-05,
|
| 883 |
+
"loss": 4.1837,
|
| 884 |
+
"step": 12500
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.12599874001259986,
|
| 888 |
+
"grad_norm": 168.18865966796875,
|
| 889 |
+
"learning_rate": 3.971779992361981e-05,
|
| 890 |
+
"loss": 3.7355,
|
| 891 |
+
"step": 12600
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.12699873001269987,
|
| 895 |
+
"grad_norm": 582.1716918945312,
|
| 896 |
+
"learning_rate": 3.971246313333807e-05,
|
| 897 |
+
"loss": 4.9486,
|
| 898 |
+
"step": 12700
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.12799872001279988,
|
| 902 |
+
"grad_norm": 337.09344482421875,
|
| 903 |
+
"learning_rate": 3.9707076716953866e-05,
|
| 904 |
+
"loss": 4.6692,
|
| 905 |
+
"step": 12800
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.12899871001289986,
|
| 909 |
+
"grad_norm": 383.8667297363281,
|
| 910 |
+
"learning_rate": 3.97016406880275e-05,
|
| 911 |
+
"loss": 3.3209,
|
| 912 |
+
"step": 12900
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.12999870001299987,
|
| 916 |
+
"grad_norm": 109.9424819946289,
|
| 917 |
+
"learning_rate": 3.9696155060244166e-05,
|
| 918 |
+
"loss": 3.5282,
|
| 919 |
+
"step": 13000
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 0.13099869001309986,
|
| 923 |
+
"grad_norm": 74.46175384521484,
|
| 924 |
+
"learning_rate": 3.969061984741393e-05,
|
| 925 |
+
"loss": 3.883,
|
| 926 |
+
"step": 13100
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.13199868001319986,
|
| 930 |
+
"grad_norm": 170.72601318359375,
|
| 931 |
+
"learning_rate": 3.9685035063471675e-05,
|
| 932 |
+
"loss": 6.4028,
|
| 933 |
+
"step": 13200
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 0.13299867001329987,
|
| 937 |
+
"grad_norm": 588.787109375,
|
| 938 |
+
"learning_rate": 3.9679400722477096e-05,
|
| 939 |
+
"loss": 3.1633,
|
| 940 |
+
"step": 13300
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 0.13399866001339986,
|
| 944 |
+
"grad_norm": 172.7151336669922,
|
| 945 |
+
"learning_rate": 3.967371683861465e-05,
|
| 946 |
+
"loss": 4.3976,
|
| 947 |
+
"step": 13400
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.13499865001349987,
|
| 951 |
+
"grad_norm": 728.732666015625,
|
| 952 |
+
"learning_rate": 3.9667983426193485e-05,
|
| 953 |
+
"loss": 3.4596,
|
| 954 |
+
"step": 13500
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.13599864001359988,
|
| 958 |
+
"grad_norm": 530.2247314453125,
|
| 959 |
+
"learning_rate": 3.9662200499647464e-05,
|
| 960 |
+
"loss": 7.6628,
|
| 961 |
+
"step": 13600
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.13699863001369986,
|
| 965 |
+
"grad_norm": 1040.433837890625,
|
| 966 |
+
"learning_rate": 3.965636807353511e-05,
|
| 967 |
+
"loss": 5.9463,
|
| 968 |
+
"step": 13700
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.13799862001379987,
|
| 972 |
+
"grad_norm": 94.73613739013672,
|
| 973 |
+
"learning_rate": 3.9650486162539555e-05,
|
| 974 |
+
"loss": 3.4241,
|
| 975 |
+
"step": 13800
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.13899861001389985,
|
| 979 |
+
"grad_norm": 87.12159729003906,
|
| 980 |
+
"learning_rate": 3.964455478146848e-05,
|
| 981 |
+
"loss": 4.2988,
|
| 982 |
+
"step": 13900
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 0.13999860001399986,
|
| 986 |
+
"grad_norm": 422.29547119140625,
|
| 987 |
+
"learning_rate": 3.963857394525413e-05,
|
| 988 |
+
"loss": 2.7699,
|
| 989 |
+
"step": 14000
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 0.14099859001409987,
|
| 993 |
+
"grad_norm": 677.4222412109375,
|
| 994 |
+
"learning_rate": 3.9632543668953284e-05,
|
| 995 |
+
"loss": 3.9013,
|
| 996 |
+
"step": 14100
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 0.14199858001419985,
|
| 1000 |
+
"grad_norm": 98.60820770263672,
|
| 1001 |
+
"learning_rate": 3.9626463967747126e-05,
|
| 1002 |
+
"loss": 2.8094,
|
| 1003 |
+
"step": 14200
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 0.14299857001429986,
|
| 1007 |
+
"grad_norm": 567.5491333007812,
|
| 1008 |
+
"learning_rate": 3.9620334856941305e-05,
|
| 1009 |
+
"loss": 1.7368,
|
| 1010 |
+
"step": 14300
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.14399856001439987,
|
| 1014 |
+
"grad_norm": 19.20868492126465,
|
| 1015 |
+
"learning_rate": 3.961415635196585e-05,
|
| 1016 |
+
"loss": 2.8347,
|
| 1017 |
+
"step": 14400
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.14499855001449985,
|
| 1021 |
+
"grad_norm": 131.01400756835938,
|
| 1022 |
+
"learning_rate": 3.960792846837514e-05,
|
| 1023 |
+
"loss": 3.9196,
|
| 1024 |
+
"step": 14500
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 0.14599854001459986,
|
| 1028 |
+
"grad_norm": 113.39109802246094,
|
| 1029 |
+
"learning_rate": 3.960165122184787e-05,
|
| 1030 |
+
"loss": 2.9107,
|
| 1031 |
+
"step": 14600
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 0.14699853001469984,
|
| 1035 |
+
"grad_norm": 701.5942993164062,
|
| 1036 |
+
"learning_rate": 3.959532462818699e-05,
|
| 1037 |
+
"loss": 1.98,
|
| 1038 |
+
"step": 14700
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 0.14799852001479985,
|
| 1042 |
+
"grad_norm": 83.20955657958984,
|
| 1043 |
+
"learning_rate": 3.958894870331971e-05,
|
| 1044 |
+
"loss": 2.093,
|
| 1045 |
+
"step": 14800
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 0.14899851001489986,
|
| 1049 |
+
"grad_norm": 473.39202880859375,
|
| 1050 |
+
"learning_rate": 3.958252346329739e-05,
|
| 1051 |
+
"loss": 3.0402,
|
| 1052 |
+
"step": 14900
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 0.14999850001499984,
|
| 1056 |
+
"grad_norm": 267.7691955566406,
|
| 1057 |
+
"learning_rate": 3.957604892429558e-05,
|
| 1058 |
+
"loss": 5.534,
|
| 1059 |
+
"step": 15000
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.15099849001509985,
|
| 1063 |
+
"grad_norm": 278.0105285644531,
|
| 1064 |
+
"learning_rate": 3.956952510261392e-05,
|
| 1065 |
+
"loss": 3.1192,
|
| 1066 |
+
"step": 15100
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.15199848001519986,
|
| 1070 |
+
"grad_norm": 76.60382080078125,
|
| 1071 |
+
"learning_rate": 3.9562952014676116e-05,
|
| 1072 |
+
"loss": 4.55,
|
| 1073 |
+
"step": 15200
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 0.15299847001529984,
|
| 1077 |
+
"grad_norm": 62.512725830078125,
|
| 1078 |
+
"learning_rate": 3.955632967702992e-05,
|
| 1079 |
+
"loss": 4.7252,
|
| 1080 |
+
"step": 15300
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 0.15399846001539985,
|
| 1084 |
+
"grad_norm": 52.81327819824219,
|
| 1085 |
+
"learning_rate": 3.954965810634706e-05,
|
| 1086 |
+
"loss": 4.3578,
|
| 1087 |
+
"step": 15400
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 0.15499845001549983,
|
| 1091 |
+
"grad_norm": 388.27850341796875,
|
| 1092 |
+
"learning_rate": 3.954293731942319e-05,
|
| 1093 |
+
"loss": 3.9787,
|
| 1094 |
+
"step": 15500
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 0.15599844001559984,
|
| 1098 |
+
"grad_norm": 83.5515365600586,
|
| 1099 |
+
"learning_rate": 3.953616733317791e-05,
|
| 1100 |
+
"loss": 0.5327,
|
| 1101 |
+
"step": 15600
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 0.15699843001569985,
|
| 1105 |
+
"grad_norm": 104.8029556274414,
|
| 1106 |
+
"learning_rate": 3.9529348164654625e-05,
|
| 1107 |
+
"loss": 5.1422,
|
| 1108 |
+
"step": 15700
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 0.15799842001579983,
|
| 1112 |
+
"grad_norm": 682.9906616210938,
|
| 1113 |
+
"learning_rate": 3.9522479831020605e-05,
|
| 1114 |
+
"loss": 0.6372,
|
| 1115 |
+
"step": 15800
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 0.15899841001589984,
|
| 1119 |
+
"grad_norm": 970.7451171875,
|
| 1120 |
+
"learning_rate": 3.951556234956686e-05,
|
| 1121 |
+
"loss": 2.5437,
|
| 1122 |
+
"step": 15900
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 0.15999840001599985,
|
| 1126 |
+
"grad_norm": 276.81207275390625,
|
| 1127 |
+
"learning_rate": 3.950859573770815e-05,
|
| 1128 |
+
"loss": 0.5738,
|
| 1129 |
+
"step": 16000
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 0.16099839001609983,
|
| 1133 |
+
"grad_norm": 121.99581909179688,
|
| 1134 |
+
"learning_rate": 3.9501580012982894e-05,
|
| 1135 |
+
"loss": 2.5115,
|
| 1136 |
+
"step": 16100
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 0.16199838001619984,
|
| 1140 |
+
"grad_norm": 234.561279296875,
|
| 1141 |
+
"learning_rate": 3.949451519305319e-05,
|
| 1142 |
+
"loss": 2.9896,
|
| 1143 |
+
"step": 16200
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 0.16299837001629983,
|
| 1147 |
+
"grad_norm": 576.3875732421875,
|
| 1148 |
+
"learning_rate": 3.948740129570471e-05,
|
| 1149 |
+
"loss": 2.834,
|
| 1150 |
+
"step": 16300
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 0.16399836001639984,
|
| 1154 |
+
"grad_norm": 256.8683166503906,
|
| 1155 |
+
"learning_rate": 3.948023833884667e-05,
|
| 1156 |
+
"loss": 2.7145,
|
| 1157 |
+
"step": 16400
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 0.16499835001649985,
|
| 1161 |
+
"grad_norm": 782.2625122070312,
|
| 1162 |
+
"learning_rate": 3.947302634051182e-05,
|
| 1163 |
+
"loss": 2.0984,
|
| 1164 |
+
"step": 16500
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 0.16599834001659983,
|
| 1168 |
+
"grad_norm": 237.81005859375,
|
| 1169 |
+
"learning_rate": 3.946576531885636e-05,
|
| 1170 |
+
"loss": 1.2758,
|
| 1171 |
+
"step": 16600
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 0.16699833001669984,
|
| 1175 |
+
"grad_norm": 148.71981811523438,
|
| 1176 |
+
"learning_rate": 3.9458455292159883e-05,
|
| 1177 |
+
"loss": 5.3672,
|
| 1178 |
+
"step": 16700
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 0.16799832001679985,
|
| 1182 |
+
"grad_norm": 339.80877685546875,
|
| 1183 |
+
"learning_rate": 3.9451096278825386e-05,
|
| 1184 |
+
"loss": 1.9513,
|
| 1185 |
+
"step": 16800
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 0.16899831001689983,
|
| 1189 |
+
"grad_norm": 339.36767578125,
|
| 1190 |
+
"learning_rate": 3.944368829737918e-05,
|
| 1191 |
+
"loss": 3.6129,
|
| 1192 |
+
"step": 16900
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 0.16999830001699984,
|
| 1196 |
+
"grad_norm": 45.14118194580078,
|
| 1197 |
+
"learning_rate": 3.9436231366470836e-05,
|
| 1198 |
+
"loss": 4.6565,
|
| 1199 |
+
"step": 17000
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 0.17099829001709982,
|
| 1203 |
+
"grad_norm": 47.00923156738281,
|
| 1204 |
+
"learning_rate": 3.942872550487318e-05,
|
| 1205 |
+
"loss": 3.9665,
|
| 1206 |
+
"step": 17100
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 0.17199828001719983,
|
| 1210 |
+
"grad_norm": 100.63198852539062,
|
| 1211 |
+
"learning_rate": 3.942117073148221e-05,
|
| 1212 |
+
"loss": 3.8538,
|
| 1213 |
+
"step": 17200
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 0.17299827001729984,
|
| 1217 |
+
"grad_norm": 13.572708129882812,
|
| 1218 |
+
"learning_rate": 3.9413567065317056e-05,
|
| 1219 |
+
"loss": 2.2025,
|
| 1220 |
+
"step": 17300
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 0.17399826001739982,
|
| 1224 |
+
"grad_norm": 411.76959228515625,
|
| 1225 |
+
"learning_rate": 3.940591452551993e-05,
|
| 1226 |
+
"loss": 1.3909,
|
| 1227 |
+
"step": 17400
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 0.17499825001749983,
|
| 1231 |
+
"grad_norm": 48.388797760009766,
|
| 1232 |
+
"learning_rate": 3.93982131313561e-05,
|
| 1233 |
+
"loss": 2.3617,
|
| 1234 |
+
"step": 17500
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 0.1759982400175998,
|
| 1238 |
+
"grad_norm": 107.87911224365234,
|
| 1239 |
+
"learning_rate": 3.939046290221383e-05,
|
| 1240 |
+
"loss": 2.1434,
|
| 1241 |
+
"step": 17600
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 0.17699823001769982,
|
| 1245 |
+
"grad_norm": 36.4320182800293,
|
| 1246 |
+
"learning_rate": 3.938266385760429e-05,
|
| 1247 |
+
"loss": 1.2506,
|
| 1248 |
+
"step": 17700
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 0.17799822001779983,
|
| 1252 |
+
"grad_norm": 293.2892150878906,
|
| 1253 |
+
"learning_rate": 3.937481601716157e-05,
|
| 1254 |
+
"loss": 3.2974,
|
| 1255 |
+
"step": 17800
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 0.1789982100178998,
|
| 1259 |
+
"grad_norm": 595.9169921875,
|
| 1260 |
+
"learning_rate": 3.936691940064261e-05,
|
| 1261 |
+
"loss": 3.2548,
|
| 1262 |
+
"step": 17900
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 0.17999820001799982,
|
| 1266 |
+
"grad_norm": 402.92236328125,
|
| 1267 |
+
"learning_rate": 3.935897402792713e-05,
|
| 1268 |
+
"loss": 1.4305,
|
| 1269 |
+
"step": 18000
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 0.18099819001809983,
|
| 1273 |
+
"grad_norm": 56.42055892944336,
|
| 1274 |
+
"learning_rate": 3.935097991901759e-05,
|
| 1275 |
+
"loss": 1.3576,
|
| 1276 |
+
"step": 18100
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 0.1819981800181998,
|
| 1280 |
+
"grad_norm": 38.25709533691406,
|
| 1281 |
+
"learning_rate": 3.934293709403915e-05,
|
| 1282 |
+
"loss": 2.9067,
|
| 1283 |
+
"step": 18200
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 0.18299817001829982,
|
| 1287 |
+
"grad_norm": 263.8206787109375,
|
| 1288 |
+
"learning_rate": 3.933484557323961e-05,
|
| 1289 |
+
"loss": 3.5546,
|
| 1290 |
+
"step": 18300
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 0.1839981600183998,
|
| 1294 |
+
"grad_norm": 190.02703857421875,
|
| 1295 |
+
"learning_rate": 3.932670537698937e-05,
|
| 1296 |
+
"loss": 3.0082,
|
| 1297 |
+
"step": 18400
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 0.1849981500184998,
|
| 1301 |
+
"grad_norm": 171.09291076660156,
|
| 1302 |
+
"learning_rate": 3.931851652578137e-05,
|
| 1303 |
+
"loss": 2.6617,
|
| 1304 |
+
"step": 18500
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 0.18599814001859982,
|
| 1308 |
+
"grad_norm": 202.73995971679688,
|
| 1309 |
+
"learning_rate": 3.931027904023102e-05,
|
| 1310 |
+
"loss": 2.818,
|
| 1311 |
+
"step": 18600
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 0.1869981300186998,
|
| 1315 |
+
"grad_norm": 179.5336456298828,
|
| 1316 |
+
"learning_rate": 3.9301992941076185e-05,
|
| 1317 |
+
"loss": 3.1686,
|
| 1318 |
+
"step": 18700
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 0.1879981200187998,
|
| 1322 |
+
"grad_norm": 396.6241455078125,
|
| 1323 |
+
"learning_rate": 3.929365824917712e-05,
|
| 1324 |
+
"loss": 1.8101,
|
| 1325 |
+
"step": 18800
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 0.18899811001889982,
|
| 1329 |
+
"grad_norm": 82.62757873535156,
|
| 1330 |
+
"learning_rate": 3.928527498551639e-05,
|
| 1331 |
+
"loss": 1.5857,
|
| 1332 |
+
"step": 18900
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 0.1899981000189998,
|
| 1336 |
+
"grad_norm": 733.2717895507812,
|
| 1337 |
+
"learning_rate": 3.9276843171198844e-05,
|
| 1338 |
+
"loss": 2.6652,
|
| 1339 |
+
"step": 19000
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 0.19099809001909981,
|
| 1343 |
+
"grad_norm": 960.3976440429688,
|
| 1344 |
+
"learning_rate": 3.926836282745158e-05,
|
| 1345 |
+
"loss": 3.8051,
|
| 1346 |
+
"step": 19100
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 0.1919980800191998,
|
| 1350 |
+
"grad_norm": 534.16748046875,
|
| 1351 |
+
"learning_rate": 3.925983397562385e-05,
|
| 1352 |
+
"loss": 2.8182,
|
| 1353 |
+
"step": 19200
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 0.1929980700192998,
|
| 1357 |
+
"grad_norm": 233.56008911132812,
|
| 1358 |
+
"learning_rate": 3.925125663718703e-05,
|
| 1359 |
+
"loss": 2.5726,
|
| 1360 |
+
"step": 19300
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 0.19399806001939982,
|
| 1364 |
+
"grad_norm": 450.2645263671875,
|
| 1365 |
+
"learning_rate": 3.924263083373455e-05,
|
| 1366 |
+
"loss": 1.7716,
|
| 1367 |
+
"step": 19400
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 0.1949980500194998,
|
| 1371 |
+
"grad_norm": 81.5189208984375,
|
| 1372 |
+
"learning_rate": 3.923395658698186e-05,
|
| 1373 |
+
"loss": 4.5006,
|
| 1374 |
+
"step": 19500
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 0.1959980400195998,
|
| 1378 |
+
"grad_norm": 84.00701141357422,
|
| 1379 |
+
"learning_rate": 3.922523391876638e-05,
|
| 1380 |
+
"loss": 0.5761,
|
| 1381 |
+
"step": 19600
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 0.19699803001969982,
|
| 1385 |
+
"grad_norm": 229.67648315429688,
|
| 1386 |
+
"learning_rate": 3.9216462851047405e-05,
|
| 1387 |
+
"loss": 5.6245,
|
| 1388 |
+
"step": 19700
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 0.1979980200197998,
|
| 1392 |
+
"grad_norm": 603.322265625,
|
| 1393 |
+
"learning_rate": 3.9207643405906094e-05,
|
| 1394 |
+
"loss": 3.7712,
|
| 1395 |
+
"step": 19800
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 0.1989980100198998,
|
| 1399 |
+
"grad_norm": 282.677734375,
|
| 1400 |
+
"learning_rate": 3.9198775605545385e-05,
|
| 1401 |
+
"loss": -0.2731,
|
| 1402 |
+
"step": 19900
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 0.1999980000199998,
|
| 1406 |
+
"grad_norm": 87.2725601196289,
|
| 1407 |
+
"learning_rate": 3.9189859472289956e-05,
|
| 1408 |
+
"loss": 2.8561,
|
| 1409 |
+
"step": 20000
|
| 1410 |
+
}
|
| 1411 |
+
],
|
| 1412 |
+
"logging_steps": 100,
|
| 1413 |
+
"max_steps": 200000,
|
| 1414 |
+
"num_input_tokens_seen": 0,
|
| 1415 |
+
"num_train_epochs": 2,
|
| 1416 |
+
"save_steps": 10000,
|
| 1417 |
+
"stateful_callbacks": {
|
| 1418 |
+
"TrainerControl": {
|
| 1419 |
+
"args": {
|
| 1420 |
+
"should_epoch_stop": false,
|
| 1421 |
+
"should_evaluate": false,
|
| 1422 |
+
"should_log": false,
|
| 1423 |
+
"should_save": true,
|
| 1424 |
+
"should_training_stop": false
|
| 1425 |
+
},
|
| 1426 |
+
"attributes": {}
|
| 1427 |
+
}
|
| 1428 |
+
},
|
| 1429 |
+
"total_flos": 0.0,
|
| 1430 |
+
"train_batch_size": 5,
|
| 1431 |
+
"trial_name": null,
|
| 1432 |
+
"trial_params": null
|
| 1433 |
+
}
|