Upload folder using huggingface_hub
Browse files- model.safetensors +1 -1
- optimizer.pt +2 -2
- rng_state.pth +3 -0
- scaler.pt +3 -0
- scheduler.pt +3 -0
- trainer_state.json +2218 -0
- training_args.bin +2 -2
model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1673955408
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ea08986330fb7c08745c35f4a1e4656cda452dadaee84515a84bbd9b582788b9
|
| 3 |
size 1673955408
|
optimizer.pt
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:b919e86d64b5a206217a3625ac0f1d51969e3950bbc7219e439328e187aa44e6
|
| 3 |
+
size 3348094138
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:655163f773de495596253784b9e225680abef2b95f5eb86ac2fcc8b632976f83
|
| 3 |
+
size 14244
|
scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ef0b5a4ee0e2737165a2c1dc88dfd59f02fd64e444dccf6a46179bda17e50d40
|
| 3 |
+
size 988
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e0f35ba3cecf286f717f8c71ec38ebb079b5ef4c3235d2082a431a3e352f81ac
|
| 3 |
+
size 1064
|
trainer_state.json
ADDED
|
@@ -0,0 +1,2218 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 1.0,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 3125,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.0032,
|
| 14 |
+
"grad_norm": 10.492237091064453,
|
| 15 |
+
"learning_rate": 1.8000000000000001e-06,
|
| 16 |
+
"loss": 4.9241,
|
| 17 |
+
"step": 10
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.0064,
|
| 21 |
+
"grad_norm": 0.9180539846420288,
|
| 22 |
+
"learning_rate": 3.8000000000000005e-06,
|
| 23 |
+
"loss": 3.0999,
|
| 24 |
+
"step": 20
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.0096,
|
| 28 |
+
"grad_norm": 0.7218229174613953,
|
| 29 |
+
"learning_rate": 5.8e-06,
|
| 30 |
+
"loss": 3.049,
|
| 31 |
+
"step": 30
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.0128,
|
| 35 |
+
"grad_norm": 0.8283277153968811,
|
| 36 |
+
"learning_rate": 7.800000000000002e-06,
|
| 37 |
+
"loss": 3.129,
|
| 38 |
+
"step": 40
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.016,
|
| 42 |
+
"grad_norm": 0.8325372338294983,
|
| 43 |
+
"learning_rate": 9.800000000000001e-06,
|
| 44 |
+
"loss": 3.1705,
|
| 45 |
+
"step": 50
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.0192,
|
| 49 |
+
"grad_norm": 0.7591913938522339,
|
| 50 |
+
"learning_rate": 1.18e-05,
|
| 51 |
+
"loss": 3.0294,
|
| 52 |
+
"step": 60
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.0224,
|
| 56 |
+
"grad_norm": 0.7909173369407654,
|
| 57 |
+
"learning_rate": 1.38e-05,
|
| 58 |
+
"loss": 2.9568,
|
| 59 |
+
"step": 70
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.0256,
|
| 63 |
+
"grad_norm": 0.8037912249565125,
|
| 64 |
+
"learning_rate": 1.58e-05,
|
| 65 |
+
"loss": 3.0724,
|
| 66 |
+
"step": 80
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.0288,
|
| 70 |
+
"grad_norm": 0.8485785722732544,
|
| 71 |
+
"learning_rate": 1.7800000000000002e-05,
|
| 72 |
+
"loss": 3.101,
|
| 73 |
+
"step": 90
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.032,
|
| 77 |
+
"grad_norm": 0.7175386548042297,
|
| 78 |
+
"learning_rate": 1.98e-05,
|
| 79 |
+
"loss": 2.9055,
|
| 80 |
+
"step": 100
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.0352,
|
| 84 |
+
"grad_norm": 0.8056895732879639,
|
| 85 |
+
"learning_rate": 1.9980592991913746e-05,
|
| 86 |
+
"loss": 2.9471,
|
| 87 |
+
"step": 110
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.0384,
|
| 91 |
+
"grad_norm": 0.9680420756340027,
|
| 92 |
+
"learning_rate": 1.995902964959569e-05,
|
| 93 |
+
"loss": 2.9588,
|
| 94 |
+
"step": 120
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.0416,
|
| 98 |
+
"grad_norm": 0.9169222712516785,
|
| 99 |
+
"learning_rate": 1.9937466307277628e-05,
|
| 100 |
+
"loss": 2.9966,
|
| 101 |
+
"step": 130
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.0448,
|
| 105 |
+
"grad_norm": 0.7769074440002441,
|
| 106 |
+
"learning_rate": 1.991590296495957e-05,
|
| 107 |
+
"loss": 2.8466,
|
| 108 |
+
"step": 140
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.048,
|
| 112 |
+
"grad_norm": 1.1656442880630493,
|
| 113 |
+
"learning_rate": 1.9894339622641512e-05,
|
| 114 |
+
"loss": 2.8495,
|
| 115 |
+
"step": 150
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.0512,
|
| 119 |
+
"grad_norm": 0.8297625780105591,
|
| 120 |
+
"learning_rate": 1.987277628032345e-05,
|
| 121 |
+
"loss": 2.7533,
|
| 122 |
+
"step": 160
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.0544,
|
| 126 |
+
"grad_norm": 0.9680063128471375,
|
| 127 |
+
"learning_rate": 1.9851212938005394e-05,
|
| 128 |
+
"loss": 2.9735,
|
| 129 |
+
"step": 170
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.0576,
|
| 133 |
+
"grad_norm": 0.8776574730873108,
|
| 134 |
+
"learning_rate": 1.9829649595687333e-05,
|
| 135 |
+
"loss": 2.8522,
|
| 136 |
+
"step": 180
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.0608,
|
| 140 |
+
"grad_norm": 0.8479920029640198,
|
| 141 |
+
"learning_rate": 1.9808086253369275e-05,
|
| 142 |
+
"loss": 2.7535,
|
| 143 |
+
"step": 190
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.064,
|
| 147 |
+
"grad_norm": 0.9027334451675415,
|
| 148 |
+
"learning_rate": 1.9786522911051214e-05,
|
| 149 |
+
"loss": 2.8398,
|
| 150 |
+
"step": 200
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.0672,
|
| 154 |
+
"grad_norm": 1.0909521579742432,
|
| 155 |
+
"learning_rate": 1.9764959568733157e-05,
|
| 156 |
+
"loss": 2.7846,
|
| 157 |
+
"step": 210
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.0704,
|
| 161 |
+
"grad_norm": 0.9581977128982544,
|
| 162 |
+
"learning_rate": 1.9743396226415095e-05,
|
| 163 |
+
"loss": 2.877,
|
| 164 |
+
"step": 220
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.0736,
|
| 168 |
+
"grad_norm": 0.8887817859649658,
|
| 169 |
+
"learning_rate": 1.9721832884097038e-05,
|
| 170 |
+
"loss": 2.7439,
|
| 171 |
+
"step": 230
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.0768,
|
| 175 |
+
"grad_norm": 1.0138421058654785,
|
| 176 |
+
"learning_rate": 1.9700269541778977e-05,
|
| 177 |
+
"loss": 2.8521,
|
| 178 |
+
"step": 240
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.08,
|
| 182 |
+
"grad_norm": 0.8431953191757202,
|
| 183 |
+
"learning_rate": 1.9678706199460916e-05,
|
| 184 |
+
"loss": 2.6156,
|
| 185 |
+
"step": 250
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.0832,
|
| 189 |
+
"grad_norm": 1.089539885520935,
|
| 190 |
+
"learning_rate": 1.9657142857142858e-05,
|
| 191 |
+
"loss": 2.7335,
|
| 192 |
+
"step": 260
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.0864,
|
| 196 |
+
"grad_norm": 0.8790264129638672,
|
| 197 |
+
"learning_rate": 1.9635579514824797e-05,
|
| 198 |
+
"loss": 2.7602,
|
| 199 |
+
"step": 270
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.0896,
|
| 203 |
+
"grad_norm": 0.942024827003479,
|
| 204 |
+
"learning_rate": 1.961401617250674e-05,
|
| 205 |
+
"loss": 2.6807,
|
| 206 |
+
"step": 280
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.0928,
|
| 210 |
+
"grad_norm": 0.9550190567970276,
|
| 211 |
+
"learning_rate": 1.9592452830188682e-05,
|
| 212 |
+
"loss": 2.7618,
|
| 213 |
+
"step": 290
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.096,
|
| 217 |
+
"grad_norm": 0.8459266424179077,
|
| 218 |
+
"learning_rate": 1.957088948787062e-05,
|
| 219 |
+
"loss": 2.6245,
|
| 220 |
+
"step": 300
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.0992,
|
| 224 |
+
"grad_norm": 0.9667879939079285,
|
| 225 |
+
"learning_rate": 1.9549326145552563e-05,
|
| 226 |
+
"loss": 2.7181,
|
| 227 |
+
"step": 310
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.1024,
|
| 231 |
+
"grad_norm": 1.0534425973892212,
|
| 232 |
+
"learning_rate": 1.9527762803234502e-05,
|
| 233 |
+
"loss": 2.6618,
|
| 234 |
+
"step": 320
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.1056,
|
| 238 |
+
"grad_norm": 0.9674198627471924,
|
| 239 |
+
"learning_rate": 1.9506199460916445e-05,
|
| 240 |
+
"loss": 2.6435,
|
| 241 |
+
"step": 330
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.1088,
|
| 245 |
+
"grad_norm": 0.972180962562561,
|
| 246 |
+
"learning_rate": 1.9484636118598384e-05,
|
| 247 |
+
"loss": 2.7582,
|
| 248 |
+
"step": 340
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.112,
|
| 252 |
+
"grad_norm": 0.8784538507461548,
|
| 253 |
+
"learning_rate": 1.9463072776280326e-05,
|
| 254 |
+
"loss": 2.654,
|
| 255 |
+
"step": 350
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.1152,
|
| 259 |
+
"grad_norm": 1.091414213180542,
|
| 260 |
+
"learning_rate": 1.944150943396227e-05,
|
| 261 |
+
"loss": 2.6457,
|
| 262 |
+
"step": 360
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.1184,
|
| 266 |
+
"grad_norm": 1.0565805435180664,
|
| 267 |
+
"learning_rate": 1.9419946091644207e-05,
|
| 268 |
+
"loss": 2.6613,
|
| 269 |
+
"step": 370
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.1216,
|
| 273 |
+
"grad_norm": 0.9876930117607117,
|
| 274 |
+
"learning_rate": 1.9398382749326146e-05,
|
| 275 |
+
"loss": 2.6771,
|
| 276 |
+
"step": 380
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.1248,
|
| 280 |
+
"grad_norm": 0.9006915092468262,
|
| 281 |
+
"learning_rate": 1.937681940700809e-05,
|
| 282 |
+
"loss": 2.6787,
|
| 283 |
+
"step": 390
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.128,
|
| 287 |
+
"grad_norm": 0.8966060876846313,
|
| 288 |
+
"learning_rate": 1.9355256064690028e-05,
|
| 289 |
+
"loss": 2.6294,
|
| 290 |
+
"step": 400
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.1312,
|
| 294 |
+
"grad_norm": 1.0113046169281006,
|
| 295 |
+
"learning_rate": 1.933369272237197e-05,
|
| 296 |
+
"loss": 2.6419,
|
| 297 |
+
"step": 410
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.1344,
|
| 301 |
+
"grad_norm": 0.914240300655365,
|
| 302 |
+
"learning_rate": 1.931212938005391e-05,
|
| 303 |
+
"loss": 2.6444,
|
| 304 |
+
"step": 420
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.1376,
|
| 308 |
+
"grad_norm": 1.0239475965499878,
|
| 309 |
+
"learning_rate": 1.929056603773585e-05,
|
| 310 |
+
"loss": 2.5037,
|
| 311 |
+
"step": 430
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.1408,
|
| 315 |
+
"grad_norm": 0.9382567405700684,
|
| 316 |
+
"learning_rate": 1.926900269541779e-05,
|
| 317 |
+
"loss": 2.7043,
|
| 318 |
+
"step": 440
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.144,
|
| 322 |
+
"grad_norm": 0.9723871350288391,
|
| 323 |
+
"learning_rate": 1.9247439353099733e-05,
|
| 324 |
+
"loss": 2.5995,
|
| 325 |
+
"step": 450
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.1472,
|
| 329 |
+
"grad_norm": 0.9572382569313049,
|
| 330 |
+
"learning_rate": 1.922587601078167e-05,
|
| 331 |
+
"loss": 2.5545,
|
| 332 |
+
"step": 460
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.1504,
|
| 336 |
+
"grad_norm": 1.187394618988037,
|
| 337 |
+
"learning_rate": 1.9204312668463614e-05,
|
| 338 |
+
"loss": 2.5126,
|
| 339 |
+
"step": 470
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.1536,
|
| 343 |
+
"grad_norm": 0.9607332348823547,
|
| 344 |
+
"learning_rate": 1.9182749326145553e-05,
|
| 345 |
+
"loss": 2.5727,
|
| 346 |
+
"step": 480
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.1568,
|
| 350 |
+
"grad_norm": 1.0382417440414429,
|
| 351 |
+
"learning_rate": 1.9161185983827495e-05,
|
| 352 |
+
"loss": 2.6179,
|
| 353 |
+
"step": 490
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.16,
|
| 357 |
+
"grad_norm": 1.0126630067825317,
|
| 358 |
+
"learning_rate": 1.9139622641509438e-05,
|
| 359 |
+
"loss": 2.6121,
|
| 360 |
+
"step": 500
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.1632,
|
| 364 |
+
"grad_norm": 1.0870510339736938,
|
| 365 |
+
"learning_rate": 1.9118059299191377e-05,
|
| 366 |
+
"loss": 2.5829,
|
| 367 |
+
"step": 510
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.1664,
|
| 371 |
+
"grad_norm": 0.9844585061073303,
|
| 372 |
+
"learning_rate": 1.909649595687332e-05,
|
| 373 |
+
"loss": 2.4699,
|
| 374 |
+
"step": 520
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.1696,
|
| 378 |
+
"grad_norm": 1.070320963859558,
|
| 379 |
+
"learning_rate": 1.9074932614555258e-05,
|
| 380 |
+
"loss": 2.5266,
|
| 381 |
+
"step": 530
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.1728,
|
| 385 |
+
"grad_norm": 0.9961358904838562,
|
| 386 |
+
"learning_rate": 1.90533692722372e-05,
|
| 387 |
+
"loss": 2.5148,
|
| 388 |
+
"step": 540
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.176,
|
| 392 |
+
"grad_norm": 0.913375198841095,
|
| 393 |
+
"learning_rate": 1.903180592991914e-05,
|
| 394 |
+
"loss": 2.4776,
|
| 395 |
+
"step": 550
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.1792,
|
| 399 |
+
"grad_norm": 0.9834626913070679,
|
| 400 |
+
"learning_rate": 1.901024258760108e-05,
|
| 401 |
+
"loss": 2.5489,
|
| 402 |
+
"step": 560
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.1824,
|
| 406 |
+
"grad_norm": 1.0209602117538452,
|
| 407 |
+
"learning_rate": 1.898867924528302e-05,
|
| 408 |
+
"loss": 2.4749,
|
| 409 |
+
"step": 570
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.1856,
|
| 413 |
+
"grad_norm": 0.9839786887168884,
|
| 414 |
+
"learning_rate": 1.896711590296496e-05,
|
| 415 |
+
"loss": 2.4735,
|
| 416 |
+
"step": 580
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.1888,
|
| 420 |
+
"grad_norm": 1.060197353363037,
|
| 421 |
+
"learning_rate": 1.8945552560646902e-05,
|
| 422 |
+
"loss": 2.5396,
|
| 423 |
+
"step": 590
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.192,
|
| 427 |
+
"grad_norm": 0.9722426533699036,
|
| 428 |
+
"learning_rate": 1.892398921832884e-05,
|
| 429 |
+
"loss": 2.4473,
|
| 430 |
+
"step": 600
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.1952,
|
| 434 |
+
"grad_norm": 1.1505547761917114,
|
| 435 |
+
"learning_rate": 1.8902425876010783e-05,
|
| 436 |
+
"loss": 2.4723,
|
| 437 |
+
"step": 610
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.1984,
|
| 441 |
+
"grad_norm": 1.0142534971237183,
|
| 442 |
+
"learning_rate": 1.8880862533692722e-05,
|
| 443 |
+
"loss": 2.4899,
|
| 444 |
+
"step": 620
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.2016,
|
| 448 |
+
"grad_norm": 1.0153712034225464,
|
| 449 |
+
"learning_rate": 1.8859299191374665e-05,
|
| 450 |
+
"loss": 2.3942,
|
| 451 |
+
"step": 630
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.2048,
|
| 455 |
+
"grad_norm": 1.0180264711380005,
|
| 456 |
+
"learning_rate": 1.8837735849056604e-05,
|
| 457 |
+
"loss": 2.389,
|
| 458 |
+
"step": 640
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.208,
|
| 462 |
+
"grad_norm": 0.9964485764503479,
|
| 463 |
+
"learning_rate": 1.8816172506738546e-05,
|
| 464 |
+
"loss": 2.407,
|
| 465 |
+
"step": 650
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.2112,
|
| 469 |
+
"grad_norm": 1.1038354635238647,
|
| 470 |
+
"learning_rate": 1.879460916442049e-05,
|
| 471 |
+
"loss": 2.4197,
|
| 472 |
+
"step": 660
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.2144,
|
| 476 |
+
"grad_norm": 1.030724287033081,
|
| 477 |
+
"learning_rate": 1.8773045822102428e-05,
|
| 478 |
+
"loss": 2.4287,
|
| 479 |
+
"step": 670
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 0.2176,
|
| 483 |
+
"grad_norm": 1.0076239109039307,
|
| 484 |
+
"learning_rate": 1.875148247978437e-05,
|
| 485 |
+
"loss": 2.4155,
|
| 486 |
+
"step": 680
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.2208,
|
| 490 |
+
"grad_norm": 1.0132572650909424,
|
| 491 |
+
"learning_rate": 1.872991913746631e-05,
|
| 492 |
+
"loss": 2.4861,
|
| 493 |
+
"step": 690
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.224,
|
| 497 |
+
"grad_norm": 0.956876814365387,
|
| 498 |
+
"learning_rate": 1.870835579514825e-05,
|
| 499 |
+
"loss": 2.4602,
|
| 500 |
+
"step": 700
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.2272,
|
| 504 |
+
"grad_norm": 1.1110647916793823,
|
| 505 |
+
"learning_rate": 1.868679245283019e-05,
|
| 506 |
+
"loss": 2.4832,
|
| 507 |
+
"step": 710
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 0.2304,
|
| 511 |
+
"grad_norm": 1.166304349899292,
|
| 512 |
+
"learning_rate": 1.866522911051213e-05,
|
| 513 |
+
"loss": 2.4793,
|
| 514 |
+
"step": 720
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.2336,
|
| 518 |
+
"grad_norm": 0.9965927600860596,
|
| 519 |
+
"learning_rate": 1.864366576819407e-05,
|
| 520 |
+
"loss": 2.4479,
|
| 521 |
+
"step": 730
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.2368,
|
| 525 |
+
"grad_norm": 1.0458307266235352,
|
| 526 |
+
"learning_rate": 1.862210242587601e-05,
|
| 527 |
+
"loss": 2.4545,
|
| 528 |
+
"step": 740
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.24,
|
| 532 |
+
"grad_norm": 1.0885543823242188,
|
| 533 |
+
"learning_rate": 1.8600539083557953e-05,
|
| 534 |
+
"loss": 2.4733,
|
| 535 |
+
"step": 750
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.2432,
|
| 539 |
+
"grad_norm": 1.083244800567627,
|
| 540 |
+
"learning_rate": 1.8578975741239892e-05,
|
| 541 |
+
"loss": 2.4689,
|
| 542 |
+
"step": 760
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.2464,
|
| 546 |
+
"grad_norm": 1.0542552471160889,
|
| 547 |
+
"learning_rate": 1.8557412398921834e-05,
|
| 548 |
+
"loss": 2.4554,
|
| 549 |
+
"step": 770
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.2496,
|
| 553 |
+
"grad_norm": 1.019744634628296,
|
| 554 |
+
"learning_rate": 1.8535849056603773e-05,
|
| 555 |
+
"loss": 2.4752,
|
| 556 |
+
"step": 780
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.2528,
|
| 560 |
+
"grad_norm": 1.078913927078247,
|
| 561 |
+
"learning_rate": 1.8514285714285716e-05,
|
| 562 |
+
"loss": 2.4116,
|
| 563 |
+
"step": 790
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.256,
|
| 567 |
+
"grad_norm": 1.0692490339279175,
|
| 568 |
+
"learning_rate": 1.8492722371967658e-05,
|
| 569 |
+
"loss": 2.4117,
|
| 570 |
+
"step": 800
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.2592,
|
| 574 |
+
"grad_norm": 1.0767900943756104,
|
| 575 |
+
"learning_rate": 1.8471159029649597e-05,
|
| 576 |
+
"loss": 2.3733,
|
| 577 |
+
"step": 810
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.2624,
|
| 581 |
+
"grad_norm": 1.1771024465560913,
|
| 582 |
+
"learning_rate": 1.844959568733154e-05,
|
| 583 |
+
"loss": 2.3885,
|
| 584 |
+
"step": 820
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.2656,
|
| 588 |
+
"grad_norm": 0.8884845972061157,
|
| 589 |
+
"learning_rate": 1.842803234501348e-05,
|
| 590 |
+
"loss": 2.359,
|
| 591 |
+
"step": 830
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 0.2688,
|
| 595 |
+
"grad_norm": 0.9761674404144287,
|
| 596 |
+
"learning_rate": 1.840646900269542e-05,
|
| 597 |
+
"loss": 2.4559,
|
| 598 |
+
"step": 840
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 0.272,
|
| 602 |
+
"grad_norm": 1.1257832050323486,
|
| 603 |
+
"learning_rate": 1.838490566037736e-05,
|
| 604 |
+
"loss": 2.4228,
|
| 605 |
+
"step": 850
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 0.2752,
|
| 609 |
+
"grad_norm": 1.009110927581787,
|
| 610 |
+
"learning_rate": 1.8363342318059302e-05,
|
| 611 |
+
"loss": 2.3719,
|
| 612 |
+
"step": 860
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 0.2784,
|
| 616 |
+
"grad_norm": 1.126349925994873,
|
| 617 |
+
"learning_rate": 1.834177897574124e-05,
|
| 618 |
+
"loss": 2.3856,
|
| 619 |
+
"step": 870
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 0.2816,
|
| 623 |
+
"grad_norm": 0.9370461106300354,
|
| 624 |
+
"learning_rate": 1.8320215633423183e-05,
|
| 625 |
+
"loss": 2.4067,
|
| 626 |
+
"step": 880
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.2848,
|
| 630 |
+
"grad_norm": 1.055823564529419,
|
| 631 |
+
"learning_rate": 1.8298652291105122e-05,
|
| 632 |
+
"loss": 2.3419,
|
| 633 |
+
"step": 890
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.288,
|
| 637 |
+
"grad_norm": 1.0766136646270752,
|
| 638 |
+
"learning_rate": 1.827708894878706e-05,
|
| 639 |
+
"loss": 2.3289,
|
| 640 |
+
"step": 900
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.2912,
|
| 644 |
+
"grad_norm": 0.9805575013160706,
|
| 645 |
+
"learning_rate": 1.8255525606469004e-05,
|
| 646 |
+
"loss": 2.3028,
|
| 647 |
+
"step": 910
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.2944,
|
| 651 |
+
"grad_norm": 1.0353854894638062,
|
| 652 |
+
"learning_rate": 1.8233962264150943e-05,
|
| 653 |
+
"loss": 2.376,
|
| 654 |
+
"step": 920
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 0.2976,
|
| 658 |
+
"grad_norm": 1.155236005783081,
|
| 659 |
+
"learning_rate": 1.8212398921832885e-05,
|
| 660 |
+
"loss": 2.4283,
|
| 661 |
+
"step": 930
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 0.3008,
|
| 665 |
+
"grad_norm": 1.1625165939331055,
|
| 666 |
+
"learning_rate": 1.8190835579514827e-05,
|
| 667 |
+
"loss": 2.2658,
|
| 668 |
+
"step": 940
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.304,
|
| 672 |
+
"grad_norm": 1.0283234119415283,
|
| 673 |
+
"learning_rate": 1.8169272237196766e-05,
|
| 674 |
+
"loss": 2.3629,
|
| 675 |
+
"step": 950
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 0.3072,
|
| 679 |
+
"grad_norm": 1.1093130111694336,
|
| 680 |
+
"learning_rate": 1.814770889487871e-05,
|
| 681 |
+
"loss": 2.3591,
|
| 682 |
+
"step": 960
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 0.3104,
|
| 686 |
+
"grad_norm": 1.1583583354949951,
|
| 687 |
+
"learning_rate": 1.8126145552560648e-05,
|
| 688 |
+
"loss": 2.398,
|
| 689 |
+
"step": 970
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.3136,
|
| 693 |
+
"grad_norm": 1.1022764444351196,
|
| 694 |
+
"learning_rate": 1.810458221024259e-05,
|
| 695 |
+
"loss": 2.3549,
|
| 696 |
+
"step": 980
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 0.3168,
|
| 700 |
+
"grad_norm": 1.0607926845550537,
|
| 701 |
+
"learning_rate": 1.808301886792453e-05,
|
| 702 |
+
"loss": 2.3334,
|
| 703 |
+
"step": 990
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 0.32,
|
| 707 |
+
"grad_norm": 1.1237353086471558,
|
| 708 |
+
"learning_rate": 1.806145552560647e-05,
|
| 709 |
+
"loss": 2.3375,
|
| 710 |
+
"step": 1000
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 0.3232,
|
| 714 |
+
"grad_norm": 1.088589072227478,
|
| 715 |
+
"learning_rate": 1.8039892183288414e-05,
|
| 716 |
+
"loss": 2.1732,
|
| 717 |
+
"step": 1010
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 0.3264,
|
| 721 |
+
"grad_norm": 1.0460518598556519,
|
| 722 |
+
"learning_rate": 1.8018328840970353e-05,
|
| 723 |
+
"loss": 2.36,
|
| 724 |
+
"step": 1020
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 0.3296,
|
| 728 |
+
"grad_norm": 1.1875877380371094,
|
| 729 |
+
"learning_rate": 1.7996765498652292e-05,
|
| 730 |
+
"loss": 2.3787,
|
| 731 |
+
"step": 1030
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 0.3328,
|
| 735 |
+
"grad_norm": 1.1336559057235718,
|
| 736 |
+
"learning_rate": 1.7975202156334234e-05,
|
| 737 |
+
"loss": 2.4228,
|
| 738 |
+
"step": 1040
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 0.336,
|
| 742 |
+
"grad_norm": 0.9651941657066345,
|
| 743 |
+
"learning_rate": 1.7953638814016173e-05,
|
| 744 |
+
"loss": 2.3414,
|
| 745 |
+
"step": 1050
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 0.3392,
|
| 749 |
+
"grad_norm": 1.1988493204116821,
|
| 750 |
+
"learning_rate": 1.7932075471698112e-05,
|
| 751 |
+
"loss": 2.3574,
|
| 752 |
+
"step": 1060
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 0.3424,
|
| 756 |
+
"grad_norm": 1.1234722137451172,
|
| 757 |
+
"learning_rate": 1.7910512129380055e-05,
|
| 758 |
+
"loss": 2.2596,
|
| 759 |
+
"step": 1070
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 0.3456,
|
| 763 |
+
"grad_norm": 1.20597505569458,
|
| 764 |
+
"learning_rate": 1.7888948787061997e-05,
|
| 765 |
+
"loss": 2.3991,
|
| 766 |
+
"step": 1080
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 0.3488,
|
| 770 |
+
"grad_norm": 1.0034183263778687,
|
| 771 |
+
"learning_rate": 1.7867385444743936e-05,
|
| 772 |
+
"loss": 2.31,
|
| 773 |
+
"step": 1090
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 0.352,
|
| 777 |
+
"grad_norm": 1.0147051811218262,
|
| 778 |
+
"learning_rate": 1.7845822102425878e-05,
|
| 779 |
+
"loss": 2.3389,
|
| 780 |
+
"step": 1100
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 0.3552,
|
| 784 |
+
"grad_norm": 1.0976855754852295,
|
| 785 |
+
"learning_rate": 1.7824258760107817e-05,
|
| 786 |
+
"loss": 2.317,
|
| 787 |
+
"step": 1110
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 0.3584,
|
| 791 |
+
"grad_norm": 1.3177474737167358,
|
| 792 |
+
"learning_rate": 1.780269541778976e-05,
|
| 793 |
+
"loss": 2.3643,
|
| 794 |
+
"step": 1120
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 0.3616,
|
| 798 |
+
"grad_norm": 1.0766491889953613,
|
| 799 |
+
"learning_rate": 1.77811320754717e-05,
|
| 800 |
+
"loss": 2.3066,
|
| 801 |
+
"step": 1130
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 0.3648,
|
| 805 |
+
"grad_norm": 1.0567567348480225,
|
| 806 |
+
"learning_rate": 1.775956873315364e-05,
|
| 807 |
+
"loss": 2.1659,
|
| 808 |
+
"step": 1140
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 0.368,
|
| 812 |
+
"grad_norm": 1.326410174369812,
|
| 813 |
+
"learning_rate": 1.7738005390835583e-05,
|
| 814 |
+
"loss": 2.2759,
|
| 815 |
+
"step": 1150
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 0.3712,
|
| 819 |
+
"grad_norm": 1.122053623199463,
|
| 820 |
+
"learning_rate": 1.7716442048517522e-05,
|
| 821 |
+
"loss": 2.342,
|
| 822 |
+
"step": 1160
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 0.3744,
|
| 826 |
+
"grad_norm": 1.052220344543457,
|
| 827 |
+
"learning_rate": 1.7694878706199465e-05,
|
| 828 |
+
"loss": 2.2605,
|
| 829 |
+
"step": 1170
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 0.3776,
|
| 833 |
+
"grad_norm": 1.1701985597610474,
|
| 834 |
+
"learning_rate": 1.7673315363881404e-05,
|
| 835 |
+
"loss": 2.3179,
|
| 836 |
+
"step": 1180
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 0.3808,
|
| 840 |
+
"grad_norm": 1.2260565757751465,
|
| 841 |
+
"learning_rate": 1.7651752021563343e-05,
|
| 842 |
+
"loss": 2.3089,
|
| 843 |
+
"step": 1190
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 0.384,
|
| 847 |
+
"grad_norm": 1.2099159955978394,
|
| 848 |
+
"learning_rate": 1.7630188679245285e-05,
|
| 849 |
+
"loss": 2.24,
|
| 850 |
+
"step": 1200
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 0.3872,
|
| 854 |
+
"grad_norm": 1.0938067436218262,
|
| 855 |
+
"learning_rate": 1.7608625336927224e-05,
|
| 856 |
+
"loss": 2.3808,
|
| 857 |
+
"step": 1210
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 0.3904,
|
| 861 |
+
"grad_norm": 1.1374958753585815,
|
| 862 |
+
"learning_rate": 1.7587061994609166e-05,
|
| 863 |
+
"loss": 2.3176,
|
| 864 |
+
"step": 1220
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 0.3936,
|
| 868 |
+
"grad_norm": 1.0443750619888306,
|
| 869 |
+
"learning_rate": 1.7565498652291105e-05,
|
| 870 |
+
"loss": 2.2128,
|
| 871 |
+
"step": 1230
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 0.3968,
|
| 875 |
+
"grad_norm": 1.0107944011688232,
|
| 876 |
+
"learning_rate": 1.7543935309973048e-05,
|
| 877 |
+
"loss": 2.3129,
|
| 878 |
+
"step": 1240
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 0.4,
|
| 882 |
+
"grad_norm": 1.1071062088012695,
|
| 883 |
+
"learning_rate": 1.7522371967654987e-05,
|
| 884 |
+
"loss": 2.2674,
|
| 885 |
+
"step": 1250
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 0.4032,
|
| 889 |
+
"grad_norm": 0.9327260851860046,
|
| 890 |
+
"learning_rate": 1.750080862533693e-05,
|
| 891 |
+
"loss": 2.2042,
|
| 892 |
+
"step": 1260
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 0.4064,
|
| 896 |
+
"grad_norm": 1.0692275762557983,
|
| 897 |
+
"learning_rate": 1.7479245283018868e-05,
|
| 898 |
+
"loss": 2.276,
|
| 899 |
+
"step": 1270
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 0.4096,
|
| 903 |
+
"grad_norm": 1.0959056615829468,
|
| 904 |
+
"learning_rate": 1.745768194070081e-05,
|
| 905 |
+
"loss": 2.3447,
|
| 906 |
+
"step": 1280
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 0.4128,
|
| 910 |
+
"grad_norm": 1.076688528060913,
|
| 911 |
+
"learning_rate": 1.743611859838275e-05,
|
| 912 |
+
"loss": 2.2936,
|
| 913 |
+
"step": 1290
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 0.416,
|
| 917 |
+
"grad_norm": 1.0170239210128784,
|
| 918 |
+
"learning_rate": 1.7414555256064692e-05,
|
| 919 |
+
"loss": 2.1855,
|
| 920 |
+
"step": 1300
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 0.4192,
|
| 924 |
+
"grad_norm": 1.1139283180236816,
|
| 925 |
+
"learning_rate": 1.7392991913746634e-05,
|
| 926 |
+
"loss": 2.2072,
|
| 927 |
+
"step": 1310
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 0.4224,
|
| 931 |
+
"grad_norm": 1.1657615900039673,
|
| 932 |
+
"learning_rate": 1.7371428571428573e-05,
|
| 933 |
+
"loss": 2.2663,
|
| 934 |
+
"step": 1320
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 0.4256,
|
| 938 |
+
"grad_norm": 1.1484967470169067,
|
| 939 |
+
"learning_rate": 1.7349865229110515e-05,
|
| 940 |
+
"loss": 2.3229,
|
| 941 |
+
"step": 1330
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 0.4288,
|
| 945 |
+
"grad_norm": 1.082829236984253,
|
| 946 |
+
"learning_rate": 1.7328301886792454e-05,
|
| 947 |
+
"loss": 2.2585,
|
| 948 |
+
"step": 1340
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 0.432,
|
| 952 |
+
"grad_norm": 1.4173678159713745,
|
| 953 |
+
"learning_rate": 1.7306738544474397e-05,
|
| 954 |
+
"loss": 2.3143,
|
| 955 |
+
"step": 1350
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 0.4352,
|
| 959 |
+
"grad_norm": 1.234764575958252,
|
| 960 |
+
"learning_rate": 1.7285175202156336e-05,
|
| 961 |
+
"loss": 2.2307,
|
| 962 |
+
"step": 1360
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 0.4384,
|
| 966 |
+
"grad_norm": 1.2216731309890747,
|
| 967 |
+
"learning_rate": 1.7263611859838275e-05,
|
| 968 |
+
"loss": 2.2255,
|
| 969 |
+
"step": 1370
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 0.4416,
|
| 973 |
+
"grad_norm": 1.0142197608947754,
|
| 974 |
+
"learning_rate": 1.7242048517520217e-05,
|
| 975 |
+
"loss": 2.2318,
|
| 976 |
+
"step": 1380
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 0.4448,
|
| 980 |
+
"grad_norm": 1.1964290142059326,
|
| 981 |
+
"learning_rate": 1.7220485175202156e-05,
|
| 982 |
+
"loss": 2.2601,
|
| 983 |
+
"step": 1390
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 0.448,
|
| 987 |
+
"grad_norm": 1.1091432571411133,
|
| 988 |
+
"learning_rate": 1.71989218328841e-05,
|
| 989 |
+
"loss": 2.2094,
|
| 990 |
+
"step": 1400
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 0.4512,
|
| 994 |
+
"grad_norm": 1.0586806535720825,
|
| 995 |
+
"learning_rate": 1.7177358490566037e-05,
|
| 996 |
+
"loss": 2.2698,
|
| 997 |
+
"step": 1410
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 0.4544,
|
| 1001 |
+
"grad_norm": 1.1744933128356934,
|
| 1002 |
+
"learning_rate": 1.715579514824798e-05,
|
| 1003 |
+
"loss": 2.1903,
|
| 1004 |
+
"step": 1420
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 0.4576,
|
| 1008 |
+
"grad_norm": 1.135746955871582,
|
| 1009 |
+
"learning_rate": 1.713423180592992e-05,
|
| 1010 |
+
"loss": 2.3062,
|
| 1011 |
+
"step": 1430
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 0.4608,
|
| 1015 |
+
"grad_norm": 1.0994842052459717,
|
| 1016 |
+
"learning_rate": 1.711266846361186e-05,
|
| 1017 |
+
"loss": 2.231,
|
| 1018 |
+
"step": 1440
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 0.464,
|
| 1022 |
+
"grad_norm": 0.9718585014343262,
|
| 1023 |
+
"learning_rate": 1.7091105121293804e-05,
|
| 1024 |
+
"loss": 2.1255,
|
| 1025 |
+
"step": 1450
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 0.4672,
|
| 1029 |
+
"grad_norm": 1.0991452932357788,
|
| 1030 |
+
"learning_rate": 1.7069541778975743e-05,
|
| 1031 |
+
"loss": 2.2461,
|
| 1032 |
+
"step": 1460
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 0.4704,
|
| 1036 |
+
"grad_norm": 1.1095904111862183,
|
| 1037 |
+
"learning_rate": 1.7047978436657685e-05,
|
| 1038 |
+
"loss": 2.1566,
|
| 1039 |
+
"step": 1470
|
| 1040 |
+
},
|
| 1041 |
+
{
|
| 1042 |
+
"epoch": 0.4736,
|
| 1043 |
+
"grad_norm": 1.0213266611099243,
|
| 1044 |
+
"learning_rate": 1.7026415094339624e-05,
|
| 1045 |
+
"loss": 2.2565,
|
| 1046 |
+
"step": 1480
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 0.4768,
|
| 1050 |
+
"grad_norm": 1.104212999343872,
|
| 1051 |
+
"learning_rate": 1.7004851752021566e-05,
|
| 1052 |
+
"loss": 2.148,
|
| 1053 |
+
"step": 1490
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 0.48,
|
| 1057 |
+
"grad_norm": 1.2078624963760376,
|
| 1058 |
+
"learning_rate": 1.6983288409703505e-05,
|
| 1059 |
+
"loss": 2.1932,
|
| 1060 |
+
"step": 1500
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 0.4832,
|
| 1064 |
+
"grad_norm": 1.095030426979065,
|
| 1065 |
+
"learning_rate": 1.6961725067385448e-05,
|
| 1066 |
+
"loss": 2.1987,
|
| 1067 |
+
"step": 1510
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 0.4864,
|
| 1071 |
+
"grad_norm": 1.2275642156600952,
|
| 1072 |
+
"learning_rate": 1.6940161725067387e-05,
|
| 1073 |
+
"loss": 2.3057,
|
| 1074 |
+
"step": 1520
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 0.4896,
|
| 1078 |
+
"grad_norm": 1.1188993453979492,
|
| 1079 |
+
"learning_rate": 1.6918598382749326e-05,
|
| 1080 |
+
"loss": 2.1914,
|
| 1081 |
+
"step": 1530
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 0.4928,
|
| 1085 |
+
"grad_norm": 1.124037504196167,
|
| 1086 |
+
"learning_rate": 1.6897035040431268e-05,
|
| 1087 |
+
"loss": 2.2604,
|
| 1088 |
+
"step": 1540
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 0.496,
|
| 1092 |
+
"grad_norm": 1.1790177822113037,
|
| 1093 |
+
"learning_rate": 1.6875471698113207e-05,
|
| 1094 |
+
"loss": 2.2696,
|
| 1095 |
+
"step": 1550
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 0.4992,
|
| 1099 |
+
"grad_norm": 1.120017409324646,
|
| 1100 |
+
"learning_rate": 1.685390835579515e-05,
|
| 1101 |
+
"loss": 2.1808,
|
| 1102 |
+
"step": 1560
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 0.5024,
|
| 1106 |
+
"grad_norm": 1.0395703315734863,
|
| 1107 |
+
"learning_rate": 1.6832345013477088e-05,
|
| 1108 |
+
"loss": 2.2044,
|
| 1109 |
+
"step": 1570
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 0.5056,
|
| 1113 |
+
"grad_norm": 1.1975572109222412,
|
| 1114 |
+
"learning_rate": 1.681078167115903e-05,
|
| 1115 |
+
"loss": 2.2375,
|
| 1116 |
+
"step": 1580
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 0.5088,
|
| 1120 |
+
"grad_norm": 1.0837329626083374,
|
| 1121 |
+
"learning_rate": 1.6789218328840973e-05,
|
| 1122 |
+
"loss": 2.2956,
|
| 1123 |
+
"step": 1590
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 0.512,
|
| 1127 |
+
"grad_norm": 1.0374228954315186,
|
| 1128 |
+
"learning_rate": 1.6767654986522912e-05,
|
| 1129 |
+
"loss": 2.1155,
|
| 1130 |
+
"step": 1600
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 0.5152,
|
| 1134 |
+
"grad_norm": 1.161590576171875,
|
| 1135 |
+
"learning_rate": 1.6746091644204854e-05,
|
| 1136 |
+
"loss": 2.2244,
|
| 1137 |
+
"step": 1610
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 0.5184,
|
| 1141 |
+
"grad_norm": 1.1977354288101196,
|
| 1142 |
+
"learning_rate": 1.6724528301886793e-05,
|
| 1143 |
+
"loss": 2.1684,
|
| 1144 |
+
"step": 1620
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 0.5216,
|
| 1148 |
+
"grad_norm": 1.2177919149398804,
|
| 1149 |
+
"learning_rate": 1.6702964959568736e-05,
|
| 1150 |
+
"loss": 2.2133,
|
| 1151 |
+
"step": 1630
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 0.5248,
|
| 1155 |
+
"grad_norm": 1.3413186073303223,
|
| 1156 |
+
"learning_rate": 1.6681401617250675e-05,
|
| 1157 |
+
"loss": 2.1853,
|
| 1158 |
+
"step": 1640
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 0.528,
|
| 1162 |
+
"grad_norm": 1.1739405393600464,
|
| 1163 |
+
"learning_rate": 1.6659838274932617e-05,
|
| 1164 |
+
"loss": 2.0678,
|
| 1165 |
+
"step": 1650
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 0.5312,
|
| 1169 |
+
"grad_norm": 1.1304874420166016,
|
| 1170 |
+
"learning_rate": 1.6638274932614556e-05,
|
| 1171 |
+
"loss": 2.2257,
|
| 1172 |
+
"step": 1660
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 0.5344,
|
| 1176 |
+
"grad_norm": 1.2166513204574585,
|
| 1177 |
+
"learning_rate": 1.66167115902965e-05,
|
| 1178 |
+
"loss": 2.1363,
|
| 1179 |
+
"step": 1670
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 0.5376,
|
| 1183 |
+
"grad_norm": 1.1125179529190063,
|
| 1184 |
+
"learning_rate": 1.6595148247978437e-05,
|
| 1185 |
+
"loss": 2.2727,
|
| 1186 |
+
"step": 1680
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 0.5408,
|
| 1190 |
+
"grad_norm": 1.147103190422058,
|
| 1191 |
+
"learning_rate": 1.657358490566038e-05,
|
| 1192 |
+
"loss": 2.0565,
|
| 1193 |
+
"step": 1690
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 0.544,
|
| 1197 |
+
"grad_norm": 1.1474051475524902,
|
| 1198 |
+
"learning_rate": 1.655202156334232e-05,
|
| 1199 |
+
"loss": 2.1802,
|
| 1200 |
+
"step": 1700
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 0.5472,
|
| 1204 |
+
"grad_norm": 1.1409813165664673,
|
| 1205 |
+
"learning_rate": 1.6530458221024258e-05,
|
| 1206 |
+
"loss": 2.1207,
|
| 1207 |
+
"step": 1710
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 0.5504,
|
| 1211 |
+
"grad_norm": 1.0063972473144531,
|
| 1212 |
+
"learning_rate": 1.65088948787062e-05,
|
| 1213 |
+
"loss": 2.1563,
|
| 1214 |
+
"step": 1720
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 0.5536,
|
| 1218 |
+
"grad_norm": 1.0541046857833862,
|
| 1219 |
+
"learning_rate": 1.6487331536388142e-05,
|
| 1220 |
+
"loss": 2.1061,
|
| 1221 |
+
"step": 1730
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 0.5568,
|
| 1225 |
+
"grad_norm": 1.1634199619293213,
|
| 1226 |
+
"learning_rate": 1.646576819407008e-05,
|
| 1227 |
+
"loss": 2.1663,
|
| 1228 |
+
"step": 1740
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 0.56,
|
| 1232 |
+
"grad_norm": 1.1393078565597534,
|
| 1233 |
+
"learning_rate": 1.6444204851752024e-05,
|
| 1234 |
+
"loss": 2.1807,
|
| 1235 |
+
"step": 1750
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 0.5632,
|
| 1239 |
+
"grad_norm": 1.1444920301437378,
|
| 1240 |
+
"learning_rate": 1.6422641509433963e-05,
|
| 1241 |
+
"loss": 2.2045,
|
| 1242 |
+
"step": 1760
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 0.5664,
|
| 1246 |
+
"grad_norm": 1.1711056232452393,
|
| 1247 |
+
"learning_rate": 1.6401078167115905e-05,
|
| 1248 |
+
"loss": 2.156,
|
| 1249 |
+
"step": 1770
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 0.5696,
|
| 1253 |
+
"grad_norm": 1.1416516304016113,
|
| 1254 |
+
"learning_rate": 1.6379514824797844e-05,
|
| 1255 |
+
"loss": 2.1471,
|
| 1256 |
+
"step": 1780
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 0.5728,
|
| 1260 |
+
"grad_norm": 1.153904676437378,
|
| 1261 |
+
"learning_rate": 1.6357951482479786e-05,
|
| 1262 |
+
"loss": 2.1123,
|
| 1263 |
+
"step": 1790
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 0.576,
|
| 1267 |
+
"grad_norm": 1.3117777109146118,
|
| 1268 |
+
"learning_rate": 1.633638814016173e-05,
|
| 1269 |
+
"loss": 2.1826,
|
| 1270 |
+
"step": 1800
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 0.5792,
|
| 1274 |
+
"grad_norm": 1.0951833724975586,
|
| 1275 |
+
"learning_rate": 1.6314824797843668e-05,
|
| 1276 |
+
"loss": 2.1882,
|
| 1277 |
+
"step": 1810
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 0.5824,
|
| 1281 |
+
"grad_norm": 1.0875611305236816,
|
| 1282 |
+
"learning_rate": 1.629326145552561e-05,
|
| 1283 |
+
"loss": 2.1407,
|
| 1284 |
+
"step": 1820
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 0.5856,
|
| 1288 |
+
"grad_norm": 1.2602307796478271,
|
| 1289 |
+
"learning_rate": 1.627169811320755e-05,
|
| 1290 |
+
"loss": 2.1009,
|
| 1291 |
+
"step": 1830
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 0.5888,
|
| 1295 |
+
"grad_norm": 1.1906789541244507,
|
| 1296 |
+
"learning_rate": 1.6250134770889488e-05,
|
| 1297 |
+
"loss": 2.217,
|
| 1298 |
+
"step": 1840
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 0.592,
|
| 1302 |
+
"grad_norm": 1.0571668148040771,
|
| 1303 |
+
"learning_rate": 1.622857142857143e-05,
|
| 1304 |
+
"loss": 2.1223,
|
| 1305 |
+
"step": 1850
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 0.5952,
|
| 1309 |
+
"grad_norm": 1.312065601348877,
|
| 1310 |
+
"learning_rate": 1.620700808625337e-05,
|
| 1311 |
+
"loss": 2.2572,
|
| 1312 |
+
"step": 1860
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 0.5984,
|
| 1316 |
+
"grad_norm": 1.1042726039886475,
|
| 1317 |
+
"learning_rate": 1.6185444743935312e-05,
|
| 1318 |
+
"loss": 2.0855,
|
| 1319 |
+
"step": 1870
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 0.6016,
|
| 1323 |
+
"grad_norm": 1.1067873239517212,
|
| 1324 |
+
"learning_rate": 1.616388140161725e-05,
|
| 1325 |
+
"loss": 2.2571,
|
| 1326 |
+
"step": 1880
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 0.6048,
|
| 1330 |
+
"grad_norm": 1.1548371315002441,
|
| 1331 |
+
"learning_rate": 1.6142318059299193e-05,
|
| 1332 |
+
"loss": 2.1176,
|
| 1333 |
+
"step": 1890
|
| 1334 |
+
},
|
| 1335 |
+
{
|
| 1336 |
+
"epoch": 0.608,
|
| 1337 |
+
"grad_norm": 1.0697139501571655,
|
| 1338 |
+
"learning_rate": 1.6120754716981132e-05,
|
| 1339 |
+
"loss": 2.2238,
|
| 1340 |
+
"step": 1900
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 0.6112,
|
| 1344 |
+
"grad_norm": 1.1137868165969849,
|
| 1345 |
+
"learning_rate": 1.6099191374663075e-05,
|
| 1346 |
+
"loss": 2.142,
|
| 1347 |
+
"step": 1910
|
| 1348 |
+
},
|
| 1349 |
+
{
|
| 1350 |
+
"epoch": 0.6144,
|
| 1351 |
+
"grad_norm": 1.082322359085083,
|
| 1352 |
+
"learning_rate": 1.6077628032345014e-05,
|
| 1353 |
+
"loss": 2.0768,
|
| 1354 |
+
"step": 1920
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 0.6176,
|
| 1358 |
+
"grad_norm": 1.0859726667404175,
|
| 1359 |
+
"learning_rate": 1.6056064690026956e-05,
|
| 1360 |
+
"loss": 2.1379,
|
| 1361 |
+
"step": 1930
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 0.6208,
|
| 1365 |
+
"grad_norm": 1.2177609205245972,
|
| 1366 |
+
"learning_rate": 1.6034501347708895e-05,
|
| 1367 |
+
"loss": 2.1137,
|
| 1368 |
+
"step": 1940
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 0.624,
|
| 1372 |
+
"grad_norm": 1.1766806840896606,
|
| 1373 |
+
"learning_rate": 1.6012938005390837e-05,
|
| 1374 |
+
"loss": 2.1459,
|
| 1375 |
+
"step": 1950
|
| 1376 |
+
},
|
| 1377 |
+
{
|
| 1378 |
+
"epoch": 0.6272,
|
| 1379 |
+
"grad_norm": 1.2313311100006104,
|
| 1380 |
+
"learning_rate": 1.599137466307278e-05,
|
| 1381 |
+
"loss": 2.1261,
|
| 1382 |
+
"step": 1960
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 0.6304,
|
| 1386 |
+
"grad_norm": 1.2768582105636597,
|
| 1387 |
+
"learning_rate": 1.596981132075472e-05,
|
| 1388 |
+
"loss": 2.1005,
|
| 1389 |
+
"step": 1970
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 0.6336,
|
| 1393 |
+
"grad_norm": 1.2875840663909912,
|
| 1394 |
+
"learning_rate": 1.594824797843666e-05,
|
| 1395 |
+
"loss": 2.0844,
|
| 1396 |
+
"step": 1980
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 0.6368,
|
| 1400 |
+
"grad_norm": 1.1797133684158325,
|
| 1401 |
+
"learning_rate": 1.59266846361186e-05,
|
| 1402 |
+
"loss": 2.071,
|
| 1403 |
+
"step": 1990
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 0.64,
|
| 1407 |
+
"grad_norm": 1.15244460105896,
|
| 1408 |
+
"learning_rate": 1.590512129380054e-05,
|
| 1409 |
+
"loss": 2.1584,
|
| 1410 |
+
"step": 2000
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"epoch": 0.6432,
|
| 1414 |
+
"grad_norm": 1.256677269935608,
|
| 1415 |
+
"learning_rate": 1.588355795148248e-05,
|
| 1416 |
+
"loss": 2.1282,
|
| 1417 |
+
"step": 2010
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 0.6464,
|
| 1421 |
+
"grad_norm": 1.1229115724563599,
|
| 1422 |
+
"learning_rate": 1.586199460916442e-05,
|
| 1423 |
+
"loss": 2.0997,
|
| 1424 |
+
"step": 2020
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 0.6496,
|
| 1428 |
+
"grad_norm": 1.2164146900177002,
|
| 1429 |
+
"learning_rate": 1.5840431266846363e-05,
|
| 1430 |
+
"loss": 2.14,
|
| 1431 |
+
"step": 2030
|
| 1432 |
+
},
|
| 1433 |
+
{
|
| 1434 |
+
"epoch": 0.6528,
|
| 1435 |
+
"grad_norm": 1.154933214187622,
|
| 1436 |
+
"learning_rate": 1.58188679245283e-05,
|
| 1437 |
+
"loss": 2.1345,
|
| 1438 |
+
"step": 2040
|
| 1439 |
+
},
|
| 1440 |
+
{
|
| 1441 |
+
"epoch": 0.656,
|
| 1442 |
+
"grad_norm": 1.182636022567749,
|
| 1443 |
+
"learning_rate": 1.5797304582210244e-05,
|
| 1444 |
+
"loss": 2.0988,
|
| 1445 |
+
"step": 2050
|
| 1446 |
+
},
|
| 1447 |
+
{
|
| 1448 |
+
"epoch": 0.6592,
|
| 1449 |
+
"grad_norm": 1.0676097869873047,
|
| 1450 |
+
"learning_rate": 1.5775741239892183e-05,
|
| 1451 |
+
"loss": 2.1429,
|
| 1452 |
+
"step": 2060
|
| 1453 |
+
},
|
| 1454 |
+
{
|
| 1455 |
+
"epoch": 0.6624,
|
| 1456 |
+
"grad_norm": 1.3782509565353394,
|
| 1457 |
+
"learning_rate": 1.5754177897574125e-05,
|
| 1458 |
+
"loss": 2.0892,
|
| 1459 |
+
"step": 2070
|
| 1460 |
+
},
|
| 1461 |
+
{
|
| 1462 |
+
"epoch": 0.6656,
|
| 1463 |
+
"grad_norm": 1.0131900310516357,
|
| 1464 |
+
"learning_rate": 1.5732614555256064e-05,
|
| 1465 |
+
"loss": 2.1193,
|
| 1466 |
+
"step": 2080
|
| 1467 |
+
},
|
| 1468 |
+
{
|
| 1469 |
+
"epoch": 0.6688,
|
| 1470 |
+
"grad_norm": 1.1614824533462524,
|
| 1471 |
+
"learning_rate": 1.5711051212938007e-05,
|
| 1472 |
+
"loss": 2.147,
|
| 1473 |
+
"step": 2090
|
| 1474 |
+
},
|
| 1475 |
+
{
|
| 1476 |
+
"epoch": 0.672,
|
| 1477 |
+
"grad_norm": 1.1127445697784424,
|
| 1478 |
+
"learning_rate": 1.568948787061995e-05,
|
| 1479 |
+
"loss": 2.1093,
|
| 1480 |
+
"step": 2100
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"epoch": 0.6752,
|
| 1484 |
+
"grad_norm": 1.1006883382797241,
|
| 1485 |
+
"learning_rate": 1.5667924528301888e-05,
|
| 1486 |
+
"loss": 2.1661,
|
| 1487 |
+
"step": 2110
|
| 1488 |
+
},
|
| 1489 |
+
{
|
| 1490 |
+
"epoch": 0.6784,
|
| 1491 |
+
"grad_norm": 1.032378077507019,
|
| 1492 |
+
"learning_rate": 1.564636118598383e-05,
|
| 1493 |
+
"loss": 2.0933,
|
| 1494 |
+
"step": 2120
|
| 1495 |
+
},
|
| 1496 |
+
{
|
| 1497 |
+
"epoch": 0.6816,
|
| 1498 |
+
"grad_norm": 1.0307894945144653,
|
| 1499 |
+
"learning_rate": 1.562479784366577e-05,
|
| 1500 |
+
"loss": 2.0407,
|
| 1501 |
+
"step": 2130
|
| 1502 |
+
},
|
| 1503 |
+
{
|
| 1504 |
+
"epoch": 0.6848,
|
| 1505 |
+
"grad_norm": 1.2763947248458862,
|
| 1506 |
+
"learning_rate": 1.5603234501347712e-05,
|
| 1507 |
+
"loss": 2.1174,
|
| 1508 |
+
"step": 2140
|
| 1509 |
+
},
|
| 1510 |
+
{
|
| 1511 |
+
"epoch": 0.688,
|
| 1512 |
+
"grad_norm": 1.2623207569122314,
|
| 1513 |
+
"learning_rate": 1.558167115902965e-05,
|
| 1514 |
+
"loss": 2.0818,
|
| 1515 |
+
"step": 2150
|
| 1516 |
+
},
|
| 1517 |
+
{
|
| 1518 |
+
"epoch": 0.6912,
|
| 1519 |
+
"grad_norm": 1.1490336656570435,
|
| 1520 |
+
"learning_rate": 1.5560107816711593e-05,
|
| 1521 |
+
"loss": 2.1214,
|
| 1522 |
+
"step": 2160
|
| 1523 |
+
},
|
| 1524 |
+
{
|
| 1525 |
+
"epoch": 0.6944,
|
| 1526 |
+
"grad_norm": 1.2585220336914062,
|
| 1527 |
+
"learning_rate": 1.5538544474393532e-05,
|
| 1528 |
+
"loss": 2.15,
|
| 1529 |
+
"step": 2170
|
| 1530 |
+
},
|
| 1531 |
+
{
|
| 1532 |
+
"epoch": 0.6976,
|
| 1533 |
+
"grad_norm": 1.203723430633545,
|
| 1534 |
+
"learning_rate": 1.551698113207547e-05,
|
| 1535 |
+
"loss": 2.1171,
|
| 1536 |
+
"step": 2180
|
| 1537 |
+
},
|
| 1538 |
+
{
|
| 1539 |
+
"epoch": 0.7008,
|
| 1540 |
+
"grad_norm": 1.3202342987060547,
|
| 1541 |
+
"learning_rate": 1.5495417789757413e-05,
|
| 1542 |
+
"loss": 2.136,
|
| 1543 |
+
"step": 2190
|
| 1544 |
+
},
|
| 1545 |
+
{
|
| 1546 |
+
"epoch": 0.704,
|
| 1547 |
+
"grad_norm": 1.2023125886917114,
|
| 1548 |
+
"learning_rate": 1.5473854447439352e-05,
|
| 1549 |
+
"loss": 2.086,
|
| 1550 |
+
"step": 2200
|
| 1551 |
+
},
|
| 1552 |
+
{
|
| 1553 |
+
"epoch": 0.7072,
|
| 1554 |
+
"grad_norm": 1.1160281896591187,
|
| 1555 |
+
"learning_rate": 1.5452291105121295e-05,
|
| 1556 |
+
"loss": 2.0867,
|
| 1557 |
+
"step": 2210
|
| 1558 |
+
},
|
| 1559 |
+
{
|
| 1560 |
+
"epoch": 0.7104,
|
| 1561 |
+
"grad_norm": 1.1412780284881592,
|
| 1562 |
+
"learning_rate": 1.5430727762803234e-05,
|
| 1563 |
+
"loss": 1.9953,
|
| 1564 |
+
"step": 2220
|
| 1565 |
+
},
|
| 1566 |
+
{
|
| 1567 |
+
"epoch": 0.7136,
|
| 1568 |
+
"grad_norm": 1.1484414339065552,
|
| 1569 |
+
"learning_rate": 1.5409164420485176e-05,
|
| 1570 |
+
"loss": 2.1114,
|
| 1571 |
+
"step": 2230
|
| 1572 |
+
},
|
| 1573 |
+
{
|
| 1574 |
+
"epoch": 0.7168,
|
| 1575 |
+
"grad_norm": 1.0628385543823242,
|
| 1576 |
+
"learning_rate": 1.538760107816712e-05,
|
| 1577 |
+
"loss": 2.0869,
|
| 1578 |
+
"step": 2240
|
| 1579 |
+
},
|
| 1580 |
+
{
|
| 1581 |
+
"epoch": 0.72,
|
| 1582 |
+
"grad_norm": 1.127213478088379,
|
| 1583 |
+
"learning_rate": 1.5366037735849057e-05,
|
| 1584 |
+
"loss": 2.169,
|
| 1585 |
+
"step": 2250
|
| 1586 |
+
},
|
| 1587 |
+
{
|
| 1588 |
+
"epoch": 0.7232,
|
| 1589 |
+
"grad_norm": 1.2877687215805054,
|
| 1590 |
+
"learning_rate": 1.5344474393531e-05,
|
| 1591 |
+
"loss": 2.1195,
|
| 1592 |
+
"step": 2260
|
| 1593 |
+
},
|
| 1594 |
+
{
|
| 1595 |
+
"epoch": 0.7264,
|
| 1596 |
+
"grad_norm": 1.2842559814453125,
|
| 1597 |
+
"learning_rate": 1.532291105121294e-05,
|
| 1598 |
+
"loss": 2.0921,
|
| 1599 |
+
"step": 2270
|
| 1600 |
+
},
|
| 1601 |
+
{
|
| 1602 |
+
"epoch": 0.7296,
|
| 1603 |
+
"grad_norm": 1.4776664972305298,
|
| 1604 |
+
"learning_rate": 1.530134770889488e-05,
|
| 1605 |
+
"loss": 2.1652,
|
| 1606 |
+
"step": 2280
|
| 1607 |
+
},
|
| 1608 |
+
{
|
| 1609 |
+
"epoch": 0.7328,
|
| 1610 |
+
"grad_norm": 1.0872628688812256,
|
| 1611 |
+
"learning_rate": 1.527978436657682e-05,
|
| 1612 |
+
"loss": 2.0732,
|
| 1613 |
+
"step": 2290
|
| 1614 |
+
},
|
| 1615 |
+
{
|
| 1616 |
+
"epoch": 0.736,
|
| 1617 |
+
"grad_norm": 1.173823356628418,
|
| 1618 |
+
"learning_rate": 1.525822102425876e-05,
|
| 1619 |
+
"loss": 2.1509,
|
| 1620 |
+
"step": 2300
|
| 1621 |
+
},
|
| 1622 |
+
{
|
| 1623 |
+
"epoch": 0.7392,
|
| 1624 |
+
"grad_norm": 1.325761318206787,
|
| 1625 |
+
"learning_rate": 1.5236657681940703e-05,
|
| 1626 |
+
"loss": 2.0753,
|
| 1627 |
+
"step": 2310
|
| 1628 |
+
},
|
| 1629 |
+
{
|
| 1630 |
+
"epoch": 0.7424,
|
| 1631 |
+
"grad_norm": 1.243660807609558,
|
| 1632 |
+
"learning_rate": 1.5215094339622642e-05,
|
| 1633 |
+
"loss": 2.071,
|
| 1634 |
+
"step": 2320
|
| 1635 |
+
},
|
| 1636 |
+
{
|
| 1637 |
+
"epoch": 0.7456,
|
| 1638 |
+
"grad_norm": 1.1181129217147827,
|
| 1639 |
+
"learning_rate": 1.5193530997304585e-05,
|
| 1640 |
+
"loss": 1.9896,
|
| 1641 |
+
"step": 2330
|
| 1642 |
+
},
|
| 1643 |
+
{
|
| 1644 |
+
"epoch": 0.7488,
|
| 1645 |
+
"grad_norm": 1.301403284072876,
|
| 1646 |
+
"learning_rate": 1.5171967654986524e-05,
|
| 1647 |
+
"loss": 2.1701,
|
| 1648 |
+
"step": 2340
|
| 1649 |
+
},
|
| 1650 |
+
{
|
| 1651 |
+
"epoch": 0.752,
|
| 1652 |
+
"grad_norm": 1.3011776208877563,
|
| 1653 |
+
"learning_rate": 1.5150404312668466e-05,
|
| 1654 |
+
"loss": 2.0567,
|
| 1655 |
+
"step": 2350
|
| 1656 |
+
},
|
| 1657 |
+
{
|
| 1658 |
+
"epoch": 0.7552,
|
| 1659 |
+
"grad_norm": 1.0706093311309814,
|
| 1660 |
+
"learning_rate": 1.5128840970350405e-05,
|
| 1661 |
+
"loss": 2.1516,
|
| 1662 |
+
"step": 2360
|
| 1663 |
+
},
|
| 1664 |
+
{
|
| 1665 |
+
"epoch": 0.7584,
|
| 1666 |
+
"grad_norm": 1.146781086921692,
|
| 1667 |
+
"learning_rate": 1.5107277628032346e-05,
|
| 1668 |
+
"loss": 2.1005,
|
| 1669 |
+
"step": 2370
|
| 1670 |
+
},
|
| 1671 |
+
{
|
| 1672 |
+
"epoch": 0.7616,
|
| 1673 |
+
"grad_norm": 1.1754928827285767,
|
| 1674 |
+
"learning_rate": 1.5085714285714288e-05,
|
| 1675 |
+
"loss": 2.0643,
|
| 1676 |
+
"step": 2380
|
| 1677 |
+
},
|
| 1678 |
+
{
|
| 1679 |
+
"epoch": 0.7648,
|
| 1680 |
+
"grad_norm": 1.1345770359039307,
|
| 1681 |
+
"learning_rate": 1.5064150943396227e-05,
|
| 1682 |
+
"loss": 2.0757,
|
| 1683 |
+
"step": 2390
|
| 1684 |
+
},
|
| 1685 |
+
{
|
| 1686 |
+
"epoch": 0.768,
|
| 1687 |
+
"grad_norm": 1.182096242904663,
|
| 1688 |
+
"learning_rate": 1.504258760107817e-05,
|
| 1689 |
+
"loss": 2.0635,
|
| 1690 |
+
"step": 2400
|
| 1691 |
+
},
|
| 1692 |
+
{
|
| 1693 |
+
"epoch": 0.7712,
|
| 1694 |
+
"grad_norm": 1.1914777755737305,
|
| 1695 |
+
"learning_rate": 1.5021024258760108e-05,
|
| 1696 |
+
"loss": 2.0608,
|
| 1697 |
+
"step": 2410
|
| 1698 |
+
},
|
| 1699 |
+
{
|
| 1700 |
+
"epoch": 0.7744,
|
| 1701 |
+
"grad_norm": 1.2629939317703247,
|
| 1702 |
+
"learning_rate": 1.499946091644205e-05,
|
| 1703 |
+
"loss": 2.0826,
|
| 1704 |
+
"step": 2420
|
| 1705 |
+
},
|
| 1706 |
+
{
|
| 1707 |
+
"epoch": 0.7776,
|
| 1708 |
+
"grad_norm": 1.2588247060775757,
|
| 1709 |
+
"learning_rate": 1.497789757412399e-05,
|
| 1710 |
+
"loss": 2.0307,
|
| 1711 |
+
"step": 2430
|
| 1712 |
+
},
|
| 1713 |
+
{
|
| 1714 |
+
"epoch": 0.7808,
|
| 1715 |
+
"grad_norm": 1.2894151210784912,
|
| 1716 |
+
"learning_rate": 1.4956334231805932e-05,
|
| 1717 |
+
"loss": 2.1513,
|
| 1718 |
+
"step": 2440
|
| 1719 |
+
},
|
| 1720 |
+
{
|
| 1721 |
+
"epoch": 0.784,
|
| 1722 |
+
"grad_norm": 1.2523419857025146,
|
| 1723 |
+
"learning_rate": 1.4934770889487873e-05,
|
| 1724 |
+
"loss": 2.1432,
|
| 1725 |
+
"step": 2450
|
| 1726 |
+
},
|
| 1727 |
+
{
|
| 1728 |
+
"epoch": 0.7872,
|
| 1729 |
+
"grad_norm": 1.167394995689392,
|
| 1730 |
+
"learning_rate": 1.4913207547169812e-05,
|
| 1731 |
+
"loss": 2.0727,
|
| 1732 |
+
"step": 2460
|
| 1733 |
+
},
|
| 1734 |
+
{
|
| 1735 |
+
"epoch": 0.7904,
|
| 1736 |
+
"grad_norm": 1.1707428693771362,
|
| 1737 |
+
"learning_rate": 1.4891644204851754e-05,
|
| 1738 |
+
"loss": 2.1085,
|
| 1739 |
+
"step": 2470
|
| 1740 |
+
},
|
| 1741 |
+
{
|
| 1742 |
+
"epoch": 0.7936,
|
| 1743 |
+
"grad_norm": 1.0771092176437378,
|
| 1744 |
+
"learning_rate": 1.4870080862533693e-05,
|
| 1745 |
+
"loss": 2.0561,
|
| 1746 |
+
"step": 2480
|
| 1747 |
+
},
|
| 1748 |
+
{
|
| 1749 |
+
"epoch": 0.7968,
|
| 1750 |
+
"grad_norm": 1.220881700515747,
|
| 1751 |
+
"learning_rate": 1.4848517520215635e-05,
|
| 1752 |
+
"loss": 2.0514,
|
| 1753 |
+
"step": 2490
|
| 1754 |
+
},
|
| 1755 |
+
{
|
| 1756 |
+
"epoch": 0.8,
|
| 1757 |
+
"grad_norm": 1.2116565704345703,
|
| 1758 |
+
"learning_rate": 1.4826954177897574e-05,
|
| 1759 |
+
"loss": 1.9991,
|
| 1760 |
+
"step": 2500
|
| 1761 |
+
},
|
| 1762 |
+
{
|
| 1763 |
+
"epoch": 0.8032,
|
| 1764 |
+
"grad_norm": 1.1328181028366089,
|
| 1765 |
+
"learning_rate": 1.4805390835579517e-05,
|
| 1766 |
+
"loss": 2.0705,
|
| 1767 |
+
"step": 2510
|
| 1768 |
+
},
|
| 1769 |
+
{
|
| 1770 |
+
"epoch": 0.8064,
|
| 1771 |
+
"grad_norm": 1.4042150974273682,
|
| 1772 |
+
"learning_rate": 1.4783827493261457e-05,
|
| 1773 |
+
"loss": 2.0447,
|
| 1774 |
+
"step": 2520
|
| 1775 |
+
},
|
| 1776 |
+
{
|
| 1777 |
+
"epoch": 0.8096,
|
| 1778 |
+
"grad_norm": 1.3431286811828613,
|
| 1779 |
+
"learning_rate": 1.4762264150943398e-05,
|
| 1780 |
+
"loss": 2.0828,
|
| 1781 |
+
"step": 2530
|
| 1782 |
+
},
|
| 1783 |
+
{
|
| 1784 |
+
"epoch": 0.8128,
|
| 1785 |
+
"grad_norm": 1.280882716178894,
|
| 1786 |
+
"learning_rate": 1.4740700808625339e-05,
|
| 1787 |
+
"loss": 1.997,
|
| 1788 |
+
"step": 2540
|
| 1789 |
+
},
|
| 1790 |
+
{
|
| 1791 |
+
"epoch": 0.816,
|
| 1792 |
+
"grad_norm": 1.3028099536895752,
|
| 1793 |
+
"learning_rate": 1.4719137466307278e-05,
|
| 1794 |
+
"loss": 2.0526,
|
| 1795 |
+
"step": 2550
|
| 1796 |
+
},
|
| 1797 |
+
{
|
| 1798 |
+
"epoch": 0.8192,
|
| 1799 |
+
"grad_norm": 1.2493317127227783,
|
| 1800 |
+
"learning_rate": 1.469757412398922e-05,
|
| 1801 |
+
"loss": 2.1792,
|
| 1802 |
+
"step": 2560
|
| 1803 |
+
},
|
| 1804 |
+
{
|
| 1805 |
+
"epoch": 0.8224,
|
| 1806 |
+
"grad_norm": 1.1215336322784424,
|
| 1807 |
+
"learning_rate": 1.4676010781671159e-05,
|
| 1808 |
+
"loss": 2.051,
|
| 1809 |
+
"step": 2570
|
| 1810 |
+
},
|
| 1811 |
+
{
|
| 1812 |
+
"epoch": 0.8256,
|
| 1813 |
+
"grad_norm": 1.222536563873291,
|
| 1814 |
+
"learning_rate": 1.4654447439353101e-05,
|
| 1815 |
+
"loss": 2.0418,
|
| 1816 |
+
"step": 2580
|
| 1817 |
+
},
|
| 1818 |
+
{
|
| 1819 |
+
"epoch": 0.8288,
|
| 1820 |
+
"grad_norm": 1.205989956855774,
|
| 1821 |
+
"learning_rate": 1.463288409703504e-05,
|
| 1822 |
+
"loss": 1.9889,
|
| 1823 |
+
"step": 2590
|
| 1824 |
+
},
|
| 1825 |
+
{
|
| 1826 |
+
"epoch": 0.832,
|
| 1827 |
+
"grad_norm": 1.1133849620819092,
|
| 1828 |
+
"learning_rate": 1.4611320754716983e-05,
|
| 1829 |
+
"loss": 2.1097,
|
| 1830 |
+
"step": 2600
|
| 1831 |
+
},
|
| 1832 |
+
{
|
| 1833 |
+
"epoch": 0.8352,
|
| 1834 |
+
"grad_norm": 1.1768999099731445,
|
| 1835 |
+
"learning_rate": 1.4589757412398923e-05,
|
| 1836 |
+
"loss": 2.1314,
|
| 1837 |
+
"step": 2610
|
| 1838 |
+
},
|
| 1839 |
+
{
|
| 1840 |
+
"epoch": 0.8384,
|
| 1841 |
+
"grad_norm": 1.1576311588287354,
|
| 1842 |
+
"learning_rate": 1.4568194070080864e-05,
|
| 1843 |
+
"loss": 2.0263,
|
| 1844 |
+
"step": 2620
|
| 1845 |
+
},
|
| 1846 |
+
{
|
| 1847 |
+
"epoch": 0.8416,
|
| 1848 |
+
"grad_norm": 1.199954867362976,
|
| 1849 |
+
"learning_rate": 1.4546630727762805e-05,
|
| 1850 |
+
"loss": 2.0265,
|
| 1851 |
+
"step": 2630
|
| 1852 |
+
},
|
| 1853 |
+
{
|
| 1854 |
+
"epoch": 0.8448,
|
| 1855 |
+
"grad_norm": 1.1806256771087646,
|
| 1856 |
+
"learning_rate": 1.4525067385444744e-05,
|
| 1857 |
+
"loss": 2.0857,
|
| 1858 |
+
"step": 2640
|
| 1859 |
+
},
|
| 1860 |
+
{
|
| 1861 |
+
"epoch": 0.848,
|
| 1862 |
+
"grad_norm": 1.3463976383209229,
|
| 1863 |
+
"learning_rate": 1.4503504043126686e-05,
|
| 1864 |
+
"loss": 2.0387,
|
| 1865 |
+
"step": 2650
|
| 1866 |
+
},
|
| 1867 |
+
{
|
| 1868 |
+
"epoch": 0.8512,
|
| 1869 |
+
"grad_norm": 1.1808844804763794,
|
| 1870 |
+
"learning_rate": 1.4481940700808625e-05,
|
| 1871 |
+
"loss": 2.0512,
|
| 1872 |
+
"step": 2660
|
| 1873 |
+
},
|
| 1874 |
+
{
|
| 1875 |
+
"epoch": 0.8544,
|
| 1876 |
+
"grad_norm": 1.3114430904388428,
|
| 1877 |
+
"learning_rate": 1.4460377358490567e-05,
|
| 1878 |
+
"loss": 2.0586,
|
| 1879 |
+
"step": 2670
|
| 1880 |
+
},
|
| 1881 |
+
{
|
| 1882 |
+
"epoch": 0.8576,
|
| 1883 |
+
"grad_norm": 1.0945067405700684,
|
| 1884 |
+
"learning_rate": 1.4438814016172508e-05,
|
| 1885 |
+
"loss": 2.0586,
|
| 1886 |
+
"step": 2680
|
| 1887 |
+
},
|
| 1888 |
+
{
|
| 1889 |
+
"epoch": 0.8608,
|
| 1890 |
+
"grad_norm": 1.3130804300308228,
|
| 1891 |
+
"learning_rate": 1.4417250673854449e-05,
|
| 1892 |
+
"loss": 2.0146,
|
| 1893 |
+
"step": 2690
|
| 1894 |
+
},
|
| 1895 |
+
{
|
| 1896 |
+
"epoch": 0.864,
|
| 1897 |
+
"grad_norm": 1.4024888277053833,
|
| 1898 |
+
"learning_rate": 1.439568733153639e-05,
|
| 1899 |
+
"loss": 2.0242,
|
| 1900 |
+
"step": 2700
|
| 1901 |
+
},
|
| 1902 |
+
{
|
| 1903 |
+
"epoch": 0.8672,
|
| 1904 |
+
"grad_norm": 1.2433137893676758,
|
| 1905 |
+
"learning_rate": 1.4374123989218328e-05,
|
| 1906 |
+
"loss": 2.0931,
|
| 1907 |
+
"step": 2710
|
| 1908 |
+
},
|
| 1909 |
+
{
|
| 1910 |
+
"epoch": 0.8704,
|
| 1911 |
+
"grad_norm": 1.1332091093063354,
|
| 1912 |
+
"learning_rate": 1.4352560646900271e-05,
|
| 1913 |
+
"loss": 2.0123,
|
| 1914 |
+
"step": 2720
|
| 1915 |
+
},
|
| 1916 |
+
{
|
| 1917 |
+
"epoch": 0.8736,
|
| 1918 |
+
"grad_norm": 1.1536661386489868,
|
| 1919 |
+
"learning_rate": 1.433099730458221e-05,
|
| 1920 |
+
"loss": 2.0524,
|
| 1921 |
+
"step": 2730
|
| 1922 |
+
},
|
| 1923 |
+
{
|
| 1924 |
+
"epoch": 0.8768,
|
| 1925 |
+
"grad_norm": 1.2748899459838867,
|
| 1926 |
+
"learning_rate": 1.4309433962264152e-05,
|
| 1927 |
+
"loss": 2.096,
|
| 1928 |
+
"step": 2740
|
| 1929 |
+
},
|
| 1930 |
+
{
|
| 1931 |
+
"epoch": 0.88,
|
| 1932 |
+
"grad_norm": 1.2594752311706543,
|
| 1933 |
+
"learning_rate": 1.4287870619946093e-05,
|
| 1934 |
+
"loss": 2.1063,
|
| 1935 |
+
"step": 2750
|
| 1936 |
+
},
|
| 1937 |
+
{
|
| 1938 |
+
"epoch": 0.8832,
|
| 1939 |
+
"grad_norm": 1.45879328250885,
|
| 1940 |
+
"learning_rate": 1.4266307277628034e-05,
|
| 1941 |
+
"loss": 2.0931,
|
| 1942 |
+
"step": 2760
|
| 1943 |
+
},
|
| 1944 |
+
{
|
| 1945 |
+
"epoch": 0.8864,
|
| 1946 |
+
"grad_norm": 1.2285680770874023,
|
| 1947 |
+
"learning_rate": 1.4244743935309974e-05,
|
| 1948 |
+
"loss": 1.9621,
|
| 1949 |
+
"step": 2770
|
| 1950 |
+
},
|
| 1951 |
+
{
|
| 1952 |
+
"epoch": 0.8896,
|
| 1953 |
+
"grad_norm": 1.201836347579956,
|
| 1954 |
+
"learning_rate": 1.4223180592991915e-05,
|
| 1955 |
+
"loss": 2.0109,
|
| 1956 |
+
"step": 2780
|
| 1957 |
+
},
|
| 1958 |
+
{
|
| 1959 |
+
"epoch": 0.8928,
|
| 1960 |
+
"grad_norm": 1.2769705057144165,
|
| 1961 |
+
"learning_rate": 1.4201617250673856e-05,
|
| 1962 |
+
"loss": 2.0142,
|
| 1963 |
+
"step": 2790
|
| 1964 |
+
},
|
| 1965 |
+
{
|
| 1966 |
+
"epoch": 0.896,
|
| 1967 |
+
"grad_norm": 1.3145017623901367,
|
| 1968 |
+
"learning_rate": 1.4180053908355795e-05,
|
| 1969 |
+
"loss": 2.0267,
|
| 1970 |
+
"step": 2800
|
| 1971 |
+
},
|
| 1972 |
+
{
|
| 1973 |
+
"epoch": 0.8992,
|
| 1974 |
+
"grad_norm": 1.1801468133926392,
|
| 1975 |
+
"learning_rate": 1.4158490566037737e-05,
|
| 1976 |
+
"loss": 2.037,
|
| 1977 |
+
"step": 2810
|
| 1978 |
+
},
|
| 1979 |
+
{
|
| 1980 |
+
"epoch": 0.9024,
|
| 1981 |
+
"grad_norm": 1.213114857673645,
|
| 1982 |
+
"learning_rate": 1.413692722371968e-05,
|
| 1983 |
+
"loss": 1.9986,
|
| 1984 |
+
"step": 2820
|
| 1985 |
+
},
|
| 1986 |
+
{
|
| 1987 |
+
"epoch": 0.9056,
|
| 1988 |
+
"grad_norm": 1.1301348209381104,
|
| 1989 |
+
"learning_rate": 1.4115363881401618e-05,
|
| 1990 |
+
"loss": 2.0354,
|
| 1991 |
+
"step": 2830
|
| 1992 |
+
},
|
| 1993 |
+
{
|
| 1994 |
+
"epoch": 0.9088,
|
| 1995 |
+
"grad_norm": 1.0965244770050049,
|
| 1996 |
+
"learning_rate": 1.4093800539083559e-05,
|
| 1997 |
+
"loss": 1.9796,
|
| 1998 |
+
"step": 2840
|
| 1999 |
+
},
|
| 2000 |
+
{
|
| 2001 |
+
"epoch": 0.912,
|
| 2002 |
+
"grad_norm": 1.2266329526901245,
|
| 2003 |
+
"learning_rate": 1.40722371967655e-05,
|
| 2004 |
+
"loss": 2.0421,
|
| 2005 |
+
"step": 2850
|
| 2006 |
+
},
|
| 2007 |
+
{
|
| 2008 |
+
"epoch": 0.9152,
|
| 2009 |
+
"grad_norm": 1.8557811975479126,
|
| 2010 |
+
"learning_rate": 1.405067385444744e-05,
|
| 2011 |
+
"loss": 2.0252,
|
| 2012 |
+
"step": 2860
|
| 2013 |
+
},
|
| 2014 |
+
{
|
| 2015 |
+
"epoch": 0.9184,
|
| 2016 |
+
"grad_norm": 1.21079683303833,
|
| 2017 |
+
"learning_rate": 1.4029110512129381e-05,
|
| 2018 |
+
"loss": 2.0597,
|
| 2019 |
+
"step": 2870
|
| 2020 |
+
},
|
| 2021 |
+
{
|
| 2022 |
+
"epoch": 0.9216,
|
| 2023 |
+
"grad_norm": 1.2704912424087524,
|
| 2024 |
+
"learning_rate": 1.4007547169811322e-05,
|
| 2025 |
+
"loss": 1.9691,
|
| 2026 |
+
"step": 2880
|
| 2027 |
+
},
|
| 2028 |
+
{
|
| 2029 |
+
"epoch": 0.9248,
|
| 2030 |
+
"grad_norm": 1.193616509437561,
|
| 2031 |
+
"learning_rate": 1.3985983827493264e-05,
|
| 2032 |
+
"loss": 2.0502,
|
| 2033 |
+
"step": 2890
|
| 2034 |
+
},
|
| 2035 |
+
{
|
| 2036 |
+
"epoch": 0.928,
|
| 2037 |
+
"grad_norm": 1.2268321514129639,
|
| 2038 |
+
"learning_rate": 1.3964420485175203e-05,
|
| 2039 |
+
"loss": 2.0497,
|
| 2040 |
+
"step": 2900
|
| 2041 |
+
},
|
| 2042 |
+
{
|
| 2043 |
+
"epoch": 0.9312,
|
| 2044 |
+
"grad_norm": 1.2767765522003174,
|
| 2045 |
+
"learning_rate": 1.3942857142857145e-05,
|
| 2046 |
+
"loss": 2.0608,
|
| 2047 |
+
"step": 2910
|
| 2048 |
+
},
|
| 2049 |
+
{
|
| 2050 |
+
"epoch": 0.9344,
|
| 2051 |
+
"grad_norm": 1.1473618745803833,
|
| 2052 |
+
"learning_rate": 1.3921293800539084e-05,
|
| 2053 |
+
"loss": 2.0007,
|
| 2054 |
+
"step": 2920
|
| 2055 |
+
},
|
| 2056 |
+
{
|
| 2057 |
+
"epoch": 0.9376,
|
| 2058 |
+
"grad_norm": 1.134504795074463,
|
| 2059 |
+
"learning_rate": 1.3899730458221025e-05,
|
| 2060 |
+
"loss": 2.0466,
|
| 2061 |
+
"step": 2930
|
| 2062 |
+
},
|
| 2063 |
+
{
|
| 2064 |
+
"epoch": 0.9408,
|
| 2065 |
+
"grad_norm": 1.0640606880187988,
|
| 2066 |
+
"learning_rate": 1.3878167115902966e-05,
|
| 2067 |
+
"loss": 1.9861,
|
| 2068 |
+
"step": 2940
|
| 2069 |
+
},
|
| 2070 |
+
{
|
| 2071 |
+
"epoch": 0.944,
|
| 2072 |
+
"grad_norm": 1.1874390840530396,
|
| 2073 |
+
"learning_rate": 1.3856603773584906e-05,
|
| 2074 |
+
"loss": 1.9654,
|
| 2075 |
+
"step": 2950
|
| 2076 |
+
},
|
| 2077 |
+
{
|
| 2078 |
+
"epoch": 0.9472,
|
| 2079 |
+
"grad_norm": 1.3798861503601074,
|
| 2080 |
+
"learning_rate": 1.3835040431266849e-05,
|
| 2081 |
+
"loss": 2.029,
|
| 2082 |
+
"step": 2960
|
| 2083 |
+
},
|
| 2084 |
+
{
|
| 2085 |
+
"epoch": 0.9504,
|
| 2086 |
+
"grad_norm": 1.1583709716796875,
|
| 2087 |
+
"learning_rate": 1.3813477088948788e-05,
|
| 2088 |
+
"loss": 1.9385,
|
| 2089 |
+
"step": 2970
|
| 2090 |
+
},
|
| 2091 |
+
{
|
| 2092 |
+
"epoch": 0.9536,
|
| 2093 |
+
"grad_norm": 1.1775461435317993,
|
| 2094 |
+
"learning_rate": 1.379191374663073e-05,
|
| 2095 |
+
"loss": 1.9918,
|
| 2096 |
+
"step": 2980
|
| 2097 |
+
},
|
| 2098 |
+
{
|
| 2099 |
+
"epoch": 0.9568,
|
| 2100 |
+
"grad_norm": 1.154227375984192,
|
| 2101 |
+
"learning_rate": 1.3770350404312669e-05,
|
| 2102 |
+
"loss": 1.9904,
|
| 2103 |
+
"step": 2990
|
| 2104 |
+
},
|
| 2105 |
+
{
|
| 2106 |
+
"epoch": 0.96,
|
| 2107 |
+
"grad_norm": 1.19357430934906,
|
| 2108 |
+
"learning_rate": 1.3748787061994611e-05,
|
| 2109 |
+
"loss": 2.0758,
|
| 2110 |
+
"step": 3000
|
| 2111 |
+
},
|
| 2112 |
+
{
|
| 2113 |
+
"epoch": 0.9632,
|
| 2114 |
+
"grad_norm": 1.256886601448059,
|
| 2115 |
+
"learning_rate": 1.372722371967655e-05,
|
| 2116 |
+
"loss": 1.9861,
|
| 2117 |
+
"step": 3010
|
| 2118 |
+
},
|
| 2119 |
+
{
|
| 2120 |
+
"epoch": 0.9664,
|
| 2121 |
+
"grad_norm": 1.2895320653915405,
|
| 2122 |
+
"learning_rate": 1.3705660377358491e-05,
|
| 2123 |
+
"loss": 2.0211,
|
| 2124 |
+
"step": 3020
|
| 2125 |
+
},
|
| 2126 |
+
{
|
| 2127 |
+
"epoch": 0.9696,
|
| 2128 |
+
"grad_norm": 1.1495587825775146,
|
| 2129 |
+
"learning_rate": 1.3684097035040433e-05,
|
| 2130 |
+
"loss": 1.977,
|
| 2131 |
+
"step": 3030
|
| 2132 |
+
},
|
| 2133 |
+
{
|
| 2134 |
+
"epoch": 0.9728,
|
| 2135 |
+
"grad_norm": 1.3089220523834229,
|
| 2136 |
+
"learning_rate": 1.3662533692722372e-05,
|
| 2137 |
+
"loss": 2.0745,
|
| 2138 |
+
"step": 3040
|
| 2139 |
+
},
|
| 2140 |
+
{
|
| 2141 |
+
"epoch": 0.976,
|
| 2142 |
+
"grad_norm": 1.1818221807479858,
|
| 2143 |
+
"learning_rate": 1.3640970350404315e-05,
|
| 2144 |
+
"loss": 2.0758,
|
| 2145 |
+
"step": 3050
|
| 2146 |
+
},
|
| 2147 |
+
{
|
| 2148 |
+
"epoch": 0.9792,
|
| 2149 |
+
"grad_norm": 1.3086206912994385,
|
| 2150 |
+
"learning_rate": 1.3619407008086254e-05,
|
| 2151 |
+
"loss": 2.134,
|
| 2152 |
+
"step": 3060
|
| 2153 |
+
},
|
| 2154 |
+
{
|
| 2155 |
+
"epoch": 0.9824,
|
| 2156 |
+
"grad_norm": 1.1708108186721802,
|
| 2157 |
+
"learning_rate": 1.3597843665768196e-05,
|
| 2158 |
+
"loss": 1.9439,
|
| 2159 |
+
"step": 3070
|
| 2160 |
+
},
|
| 2161 |
+
{
|
| 2162 |
+
"epoch": 0.9856,
|
| 2163 |
+
"grad_norm": 1.1925756931304932,
|
| 2164 |
+
"learning_rate": 1.3576280323450135e-05,
|
| 2165 |
+
"loss": 2.0295,
|
| 2166 |
+
"step": 3080
|
| 2167 |
+
},
|
| 2168 |
+
{
|
| 2169 |
+
"epoch": 0.9888,
|
| 2170 |
+
"grad_norm": 1.1521224975585938,
|
| 2171 |
+
"learning_rate": 1.3554716981132078e-05,
|
| 2172 |
+
"loss": 1.9799,
|
| 2173 |
+
"step": 3090
|
| 2174 |
+
},
|
| 2175 |
+
{
|
| 2176 |
+
"epoch": 0.992,
|
| 2177 |
+
"grad_norm": 1.3074902296066284,
|
| 2178 |
+
"learning_rate": 1.3533153638814018e-05,
|
| 2179 |
+
"loss": 2.0069,
|
| 2180 |
+
"step": 3100
|
| 2181 |
+
},
|
| 2182 |
+
{
|
| 2183 |
+
"epoch": 0.9952,
|
| 2184 |
+
"grad_norm": 1.3151848316192627,
|
| 2185 |
+
"learning_rate": 1.3511590296495957e-05,
|
| 2186 |
+
"loss": 1.9447,
|
| 2187 |
+
"step": 3110
|
| 2188 |
+
},
|
| 2189 |
+
{
|
| 2190 |
+
"epoch": 0.9984,
|
| 2191 |
+
"grad_norm": 1.2995407581329346,
|
| 2192 |
+
"learning_rate": 1.34900269541779e-05,
|
| 2193 |
+
"loss": 1.971,
|
| 2194 |
+
"step": 3120
|
| 2195 |
+
}
|
| 2196 |
+
],
|
| 2197 |
+
"logging_steps": 10,
|
| 2198 |
+
"max_steps": 9375,
|
| 2199 |
+
"num_input_tokens_seen": 0,
|
| 2200 |
+
"num_train_epochs": 3,
|
| 2201 |
+
"save_steps": 500,
|
| 2202 |
+
"stateful_callbacks": {
|
| 2203 |
+
"TrainerControl": {
|
| 2204 |
+
"args": {
|
| 2205 |
+
"should_epoch_stop": false,
|
| 2206 |
+
"should_evaluate": false,
|
| 2207 |
+
"should_log": false,
|
| 2208 |
+
"should_save": true,
|
| 2209 |
+
"should_training_stop": false
|
| 2210 |
+
},
|
| 2211 |
+
"attributes": {}
|
| 2212 |
+
}
|
| 2213 |
+
},
|
| 2214 |
+
"total_flos": 1.11089319936e+17,
|
| 2215 |
+
"train_batch_size": 8,
|
| 2216 |
+
"trial_name": null,
|
| 2217 |
+
"trial_params": null
|
| 2218 |
+
}
|
training_args.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 sha256:58957521ddf4130592fd11e650e23deecae4ae1e25186f43a944260f93e76159
|
| 3 |
+
size 5240
|