Commit
•
12e4fbc
1
Parent(s):
aa5c8aa
Upload folder using huggingface_hub (#1)
Browse files- 2b445798853676171dbb3172edd48d582d7479694914ad7d7ff176bf0ed5fc43 (2d5cb38c6fd4ae7ae412bcceade9e33c3a079f25)
- model.safetensors +3 -0
- optimizer.pt +3 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- trainer_state.json +1585 -0
- training_args.bin +3 -0
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0fc3a645a790262e5fff8923eb30cb75bc640a05333b8348b23ff05940be04b1
|
| 3 |
+
size 379087640
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2172b035c05b1f601b9237d2e6b3321bf4c57e3204baea57989e7d077bc8106e
|
| 3 |
+
size 3152480
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f8c88c483589e25c71a3212244d6755ea974f610775bad36d935b3df14dc617d
|
| 3 |
+
size 14244
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5d8c82186737c1208be5544e2a6dfb5eaf7f4a9ade52220334c78a811e5286d6
|
| 3 |
+
size 1000
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1585 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.12010584022747946,
|
| 3 |
+
"best_model_checkpoint": "/workspace/disk2/krishna/checkpoints/checkpoint-940",
|
| 4 |
+
"epoch": 0.097,
|
| 5 |
+
"eval_steps": 10,
|
| 6 |
+
"global_step": 970,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.001,
|
| 13 |
+
"grad_norm": 0.0356324203312397,
|
| 14 |
+
"learning_rate": 1e-05,
|
| 15 |
+
"loss": 0.1207,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.001,
|
| 20 |
+
"eval_cos_sim": 0.8792359232902527,
|
| 21 |
+
"eval_loss": 0.12173220255124045,
|
| 22 |
+
"eval_runtime": 171.7728,
|
| 23 |
+
"eval_samples_per_second": 23.287,
|
| 24 |
+
"eval_steps_per_second": 0.367,
|
| 25 |
+
"step": 10
|
| 26 |
+
},
|
| 27 |
+
{
|
| 28 |
+
"epoch": 0.002,
|
| 29 |
+
"grad_norm": 0.030419372022151947,
|
| 30 |
+
"learning_rate": 2e-05,
|
| 31 |
+
"loss": 0.1213,
|
| 32 |
+
"step": 20
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.002,
|
| 36 |
+
"eval_cos_sim": 0.8789854049682617,
|
| 37 |
+
"eval_loss": 0.12198310520398092,
|
| 38 |
+
"eval_runtime": 159.1521,
|
| 39 |
+
"eval_samples_per_second": 25.133,
|
| 40 |
+
"eval_steps_per_second": 0.396,
|
| 41 |
+
"step": 20
|
| 42 |
+
},
|
| 43 |
+
{
|
| 44 |
+
"epoch": 0.003,
|
| 45 |
+
"grad_norm": 0.033041320741176605,
|
| 46 |
+
"learning_rate": 3e-05,
|
| 47 |
+
"loss": 0.1204,
|
| 48 |
+
"step": 30
|
| 49 |
+
},
|
| 50 |
+
{
|
| 51 |
+
"epoch": 0.003,
|
| 52 |
+
"eval_cos_sim": 0.8790653347969055,
|
| 53 |
+
"eval_loss": 0.12189955379712057,
|
| 54 |
+
"eval_runtime": 161.0741,
|
| 55 |
+
"eval_samples_per_second": 24.833,
|
| 56 |
+
"eval_steps_per_second": 0.391,
|
| 57 |
+
"step": 30
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 0.004,
|
| 61 |
+
"grad_norm": 0.04209210351109505,
|
| 62 |
+
"learning_rate": 4e-05,
|
| 63 |
+
"loss": 0.1213,
|
| 64 |
+
"step": 40
|
| 65 |
+
},
|
| 66 |
+
{
|
| 67 |
+
"epoch": 0.004,
|
| 68 |
+
"eval_cos_sim": 0.8792155385017395,
|
| 69 |
+
"eval_loss": 0.12175503399121237,
|
| 70 |
+
"eval_runtime": 159.4228,
|
| 71 |
+
"eval_samples_per_second": 25.091,
|
| 72 |
+
"eval_steps_per_second": 0.395,
|
| 73 |
+
"step": 40
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.005,
|
| 77 |
+
"grad_norm": 0.03182140365242958,
|
| 78 |
+
"learning_rate": 5e-05,
|
| 79 |
+
"loss": 0.1203,
|
| 80 |
+
"step": 50
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.005,
|
| 84 |
+
"eval_cos_sim": 0.8791962265968323,
|
| 85 |
+
"eval_loss": 0.12176946785199118,
|
| 86 |
+
"eval_runtime": 160.3207,
|
| 87 |
+
"eval_samples_per_second": 24.95,
|
| 88 |
+
"eval_steps_per_second": 0.393,
|
| 89 |
+
"step": 50
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 0.006,
|
| 93 |
+
"grad_norm": 0.05823719501495361,
|
| 94 |
+
"learning_rate": 2.4802665827257164e-05,
|
| 95 |
+
"loss": 0.1213,
|
| 96 |
+
"step": 60
|
| 97 |
+
},
|
| 98 |
+
{
|
| 99 |
+
"epoch": 0.006,
|
| 100 |
+
"eval_cos_sim": 0.8791635036468506,
|
| 101 |
+
"eval_loss": 0.12179789688336325,
|
| 102 |
+
"eval_runtime": 164.7575,
|
| 103 |
+
"eval_samples_per_second": 24.278,
|
| 104 |
+
"eval_steps_per_second": 0.382,
|
| 105 |
+
"step": 60
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 0.007,
|
| 109 |
+
"grad_norm": 0.02305755950510502,
|
| 110 |
+
"learning_rate": 4.999688473794144e-05,
|
| 111 |
+
"loss": 0.1211,
|
| 112 |
+
"step": 70
|
| 113 |
+
},
|
| 114 |
+
{
|
| 115 |
+
"epoch": 0.007,
|
| 116 |
+
"eval_cos_sim": 0.8792662024497986,
|
| 117 |
+
"eval_loss": 0.12169700815426779,
|
| 118 |
+
"eval_runtime": 159.8158,
|
| 119 |
+
"eval_samples_per_second": 25.029,
|
| 120 |
+
"eval_steps_per_second": 0.394,
|
| 121 |
+
"step": 70
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.008,
|
| 125 |
+
"grad_norm": 0.03906348720192909,
|
| 126 |
+
"learning_rate": 2.4408046661584414e-05,
|
| 127 |
+
"loss": 0.1201,
|
| 128 |
+
"step": 80
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.008,
|
| 132 |
+
"eval_cos_sim": 0.879157543182373,
|
| 133 |
+
"eval_loss": 0.12181253530728293,
|
| 134 |
+
"eval_runtime": 162.3324,
|
| 135 |
+
"eval_samples_per_second": 24.641,
|
| 136 |
+
"eval_steps_per_second": 0.388,
|
| 137 |
+
"step": 80
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 0.009,
|
| 141 |
+
"grad_norm": 0.0275803804397583,
|
| 142 |
+
"learning_rate": 4.998753972815434e-05,
|
| 143 |
+
"loss": 0.1208,
|
| 144 |
+
"step": 90
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 0.009,
|
| 148 |
+
"eval_cos_sim": 0.8792516589164734,
|
| 149 |
+
"eval_loss": 0.121718086264009,
|
| 150 |
+
"eval_runtime": 158.5035,
|
| 151 |
+
"eval_samples_per_second": 25.236,
|
| 152 |
+
"eval_steps_per_second": 0.397,
|
| 153 |
+
"step": 90
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 0.01,
|
| 157 |
+
"grad_norm": 0.03248042240738869,
|
| 158 |
+
"learning_rate": 2.4013575023093667e-05,
|
| 159 |
+
"loss": 0.1224,
|
| 160 |
+
"step": 100
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"epoch": 0.01,
|
| 164 |
+
"eval_cos_sim": 0.879462718963623,
|
| 165 |
+
"eval_loss": 0.1215014844154067,
|
| 166 |
+
"eval_runtime": 159.5892,
|
| 167 |
+
"eval_samples_per_second": 25.064,
|
| 168 |
+
"eval_steps_per_second": 0.395,
|
| 169 |
+
"step": 100
|
| 170 |
+
},
|
| 171 |
+
{
|
| 172 |
+
"epoch": 0.011,
|
| 173 |
+
"grad_norm": 0.03436814621090889,
|
| 174 |
+
"learning_rate": 4.9971967299611097e-05,
|
| 175 |
+
"loss": 0.1205,
|
| 176 |
+
"step": 110
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 0.011,
|
| 180 |
+
"eval_cos_sim": 0.8795427680015564,
|
| 181 |
+
"eval_loss": 0.1214234667037673,
|
| 182 |
+
"eval_runtime": 168.2892,
|
| 183 |
+
"eval_samples_per_second": 23.769,
|
| 184 |
+
"eval_steps_per_second": 0.374,
|
| 185 |
+
"step": 110
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.012,
|
| 189 |
+
"grad_norm": 0.03663235530257225,
|
| 190 |
+
"learning_rate": 2.3619349222387287e-05,
|
| 191 |
+
"loss": 0.1209,
|
| 192 |
+
"step": 120
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.012,
|
| 196 |
+
"eval_cos_sim": 0.8793898224830627,
|
| 197 |
+
"eval_loss": 0.1215821056579299,
|
| 198 |
+
"eval_runtime": 170.7269,
|
| 199 |
+
"eval_samples_per_second": 23.429,
|
| 200 |
+
"eval_steps_per_second": 0.369,
|
| 201 |
+
"step": 120
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 0.013,
|
| 205 |
+
"grad_norm": 0.03549114614725113,
|
| 206 |
+
"learning_rate": 4.9950171333287335e-05,
|
| 207 |
+
"loss": 0.1218,
|
| 208 |
+
"step": 130
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 0.013,
|
| 212 |
+
"eval_cos_sim": 0.8795300722122192,
|
| 213 |
+
"eval_loss": 0.12144066002118063,
|
| 214 |
+
"eval_runtime": 162.7257,
|
| 215 |
+
"eval_samples_per_second": 24.581,
|
| 216 |
+
"eval_steps_per_second": 0.387,
|
| 217 |
+
"step": 130
|
| 218 |
+
},
|
| 219 |
+
{
|
| 220 |
+
"epoch": 0.014,
|
| 221 |
+
"grad_norm": 0.03164505586028099,
|
| 222 |
+
"learning_rate": 2.3225467508799633e-05,
|
| 223 |
+
"loss": 0.1208,
|
| 224 |
+
"step": 140
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 0.014,
|
| 228 |
+
"eval_cos_sim": 0.8797659873962402,
|
| 229 |
+
"eval_loss": 0.12119961311566306,
|
| 230 |
+
"eval_runtime": 163.5115,
|
| 231 |
+
"eval_samples_per_second": 24.463,
|
| 232 |
+
"eval_steps_per_second": 0.385,
|
| 233 |
+
"step": 140
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.015,
|
| 237 |
+
"grad_norm": 0.031108180060982704,
|
| 238 |
+
"learning_rate": 4.992215726119483e-05,
|
| 239 |
+
"loss": 0.1213,
|
| 240 |
+
"step": 150
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.015,
|
| 244 |
+
"eval_cos_sim": 0.8797353506088257,
|
| 245 |
+
"eval_loss": 0.1212306742881484,
|
| 246 |
+
"eval_runtime": 165.866,
|
| 247 |
+
"eval_samples_per_second": 24.116,
|
| 248 |
+
"eval_steps_per_second": 0.38,
|
| 249 |
+
"step": 150
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 0.016,
|
| 253 |
+
"grad_norm": 0.030103642493486404,
|
| 254 |
+
"learning_rate": 2.2832028045911203e-05,
|
| 255 |
+
"loss": 0.1209,
|
| 256 |
+
"step": 160
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 0.016,
|
| 260 |
+
"eval_cos_sim": 0.8793777823448181,
|
| 261 |
+
"eval_loss": 0.12159336998211813,
|
| 262 |
+
"eval_runtime": 171.7298,
|
| 263 |
+
"eval_samples_per_second": 23.292,
|
| 264 |
+
"eval_steps_per_second": 0.367,
|
| 265 |
+
"step": 160
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 0.017,
|
| 269 |
+
"grad_norm": 0.05055614188313484,
|
| 270 |
+
"learning_rate": 4.9887932065027656e-05,
|
| 271 |
+
"loss": 0.1204,
|
| 272 |
+
"step": 170
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 0.017,
|
| 276 |
+
"eval_cos_sim": 0.8795183300971985,
|
| 277 |
+
"eval_loss": 0.1214520344947524,
|
| 278 |
+
"eval_runtime": 162.2461,
|
| 279 |
+
"eval_samples_per_second": 24.654,
|
| 280 |
+
"eval_steps_per_second": 0.388,
|
| 281 |
+
"step": 170
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 0.018,
|
| 285 |
+
"grad_norm": 0.03837039694190025,
|
| 286 |
+
"learning_rate": 2.2439128887084646e-05,
|
| 287 |
+
"loss": 0.1202,
|
| 288 |
+
"step": 180
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 0.018,
|
| 292 |
+
"eval_cos_sim": 0.8797397017478943,
|
| 293 |
+
"eval_loss": 0.12122445728527975,
|
| 294 |
+
"eval_runtime": 161.451,
|
| 295 |
+
"eval_samples_per_second": 24.775,
|
| 296 |
+
"eval_steps_per_second": 0.39,
|
| 297 |
+
"step": 180
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.019,
|
| 301 |
+
"grad_norm": 0.03563898801803589,
|
| 302 |
+
"learning_rate": 4.98475042744222e-05,
|
| 303 |
+
"loss": 0.1221,
|
| 304 |
+
"step": 190
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.019,
|
| 308 |
+
"eval_cos_sim": 0.8797785639762878,
|
| 309 |
+
"eval_loss": 0.12118578769909812,
|
| 310 |
+
"eval_runtime": 157.9064,
|
| 311 |
+
"eval_samples_per_second": 25.331,
|
| 312 |
+
"eval_steps_per_second": 0.399,
|
| 313 |
+
"step": 190
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 0.02,
|
| 317 |
+
"grad_norm": 0.0392858162522316,
|
| 318 |
+
"learning_rate": 2.204686795102736e-05,
|
| 319 |
+
"loss": 0.1204,
|
| 320 |
+
"step": 200
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 0.02,
|
| 324 |
+
"eval_cos_sim": 0.8796395063400269,
|
| 325 |
+
"eval_loss": 0.12133027555691672,
|
| 326 |
+
"eval_runtime": 163.7884,
|
| 327 |
+
"eval_samples_per_second": 24.422,
|
| 328 |
+
"eval_steps_per_second": 0.385,
|
| 329 |
+
"step": 200
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 0.021,
|
| 333 |
+
"grad_norm": 0.04556349664926529,
|
| 334 |
+
"learning_rate": 4.980088396483144e-05,
|
| 335 |
+
"loss": 0.1205,
|
| 336 |
+
"step": 210
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 0.021,
|
| 340 |
+
"eval_cos_sim": 0.8796445727348328,
|
| 341 |
+
"eval_loss": 0.12132433941113424,
|
| 342 |
+
"eval_runtime": 164.6589,
|
| 343 |
+
"eval_samples_per_second": 24.293,
|
| 344 |
+
"eval_steps_per_second": 0.383,
|
| 345 |
+
"step": 210
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.022,
|
| 349 |
+
"grad_norm": 0.030130930244922638,
|
| 350 |
+
"learning_rate": 2.1655342997387947e-05,
|
| 351 |
+
"loss": 0.1201,
|
| 352 |
+
"step": 220
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.022,
|
| 356 |
+
"eval_cos_sim": 0.8796879649162292,
|
| 357 |
+
"eval_loss": 0.12127337600933981,
|
| 358 |
+
"eval_runtime": 163.052,
|
| 359 |
+
"eval_samples_per_second": 24.532,
|
| 360 |
+
"eval_steps_per_second": 0.386,
|
| 361 |
+
"step": 220
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 0.023,
|
| 365 |
+
"grad_norm": 0.027453621849417686,
|
| 366 |
+
"learning_rate": 4.9748082755013934e-05,
|
| 367 |
+
"loss": 0.1205,
|
| 368 |
+
"step": 230
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 0.023,
|
| 372 |
+
"eval_cos_sim": 0.8797481060028076,
|
| 373 |
+
"eval_loss": 0.12121248771893454,
|
| 374 |
+
"eval_runtime": 159.6988,
|
| 375 |
+
"eval_samples_per_second": 25.047,
|
| 376 |
+
"eval_steps_per_second": 0.394,
|
| 377 |
+
"step": 230
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 0.024,
|
| 381 |
+
"grad_norm": 0.029768602922558784,
|
| 382 |
+
"learning_rate": 2.126465160239341e-05,
|
| 383 |
+
"loss": 0.1206,
|
| 384 |
+
"step": 240
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 0.024,
|
| 388 |
+
"eval_cos_sim": 0.8797679543495178,
|
| 389 |
+
"eval_loss": 0.12119435026394797,
|
| 390 |
+
"eval_runtime": 170.5172,
|
| 391 |
+
"eval_samples_per_second": 23.458,
|
| 392 |
+
"eval_steps_per_second": 0.369,
|
| 393 |
+
"step": 240
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 0.025,
|
| 397 |
+
"grad_norm": 0.025975426658988,
|
| 398 |
+
"learning_rate": 4.968911380413809e-05,
|
| 399 |
+
"loss": 0.1206,
|
| 400 |
+
"step": 250
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 0.025,
|
| 404 |
+
"eval_cos_sim": 0.8798050284385681,
|
| 405 |
+
"eval_loss": 0.12115855839001609,
|
| 406 |
+
"eval_runtime": 162.5635,
|
| 407 |
+
"eval_samples_per_second": 24.606,
|
| 408 |
+
"eval_steps_per_second": 0.388,
|
| 409 |
+
"step": 250
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.026,
|
| 413 |
+
"grad_norm": 0.032136961817741394,
|
| 414 |
+
"learning_rate": 2.0874891134530094e-05,
|
| 415 |
+
"loss": 0.1207,
|
| 416 |
+
"step": 260
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.026,
|
| 420 |
+
"eval_cos_sim": 0.8799233436584473,
|
| 421 |
+
"eval_loss": 0.12104250910031271,
|
| 422 |
+
"eval_runtime": 171.0657,
|
| 423 |
+
"eval_samples_per_second": 23.383,
|
| 424 |
+
"eval_steps_per_second": 0.368,
|
| 425 |
+
"step": 260
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 0.027,
|
| 429 |
+
"grad_norm": 0.035989198833703995,
|
| 430 |
+
"learning_rate": 4.962399180850275e-05,
|
| 431 |
+
"loss": 0.12,
|
| 432 |
+
"step": 270
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"epoch": 0.027,
|
| 436 |
+
"eval_cos_sim": 0.8800029754638672,
|
| 437 |
+
"eval_loss": 0.12096367742764426,
|
| 438 |
+
"eval_runtime": 162.5704,
|
| 439 |
+
"eval_samples_per_second": 24.605,
|
| 440 |
+
"eval_steps_per_second": 0.388,
|
| 441 |
+
"step": 270
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 0.028,
|
| 445 |
+
"grad_norm": 0.02917526848614216,
|
| 446 |
+
"learning_rate": 2.0486158730277393e-05,
|
| 447 |
+
"loss": 0.1205,
|
| 448 |
+
"step": 280
|
| 449 |
+
},
|
| 450 |
+
{
|
| 451 |
+
"epoch": 0.028,
|
| 452 |
+
"eval_cos_sim": 0.8800209164619446,
|
| 453 |
+
"eval_loss": 0.12094438698040914,
|
| 454 |
+
"eval_runtime": 163.135,
|
| 455 |
+
"eval_samples_per_second": 24.52,
|
| 456 |
+
"eval_steps_per_second": 0.386,
|
| 457 |
+
"step": 280
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.029,
|
| 461 |
+
"grad_norm": 0.040587518364191055,
|
| 462 |
+
"learning_rate": 4.955273299787453e-05,
|
| 463 |
+
"loss": 0.1204,
|
| 464 |
+
"step": 290
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.029,
|
| 468 |
+
"eval_cos_sim": 0.8800665140151978,
|
| 469 |
+
"eval_loss": 0.12090009071576072,
|
| 470 |
+
"eval_runtime": 160.8422,
|
| 471 |
+
"eval_samples_per_second": 24.869,
|
| 472 |
+
"eval_steps_per_second": 0.392,
|
| 473 |
+
"step": 290
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 0.03,
|
| 477 |
+
"grad_norm": 0.02535935305058956,
|
| 478 |
+
"learning_rate": 2.00985512699005e-05,
|
| 479 |
+
"loss": 0.121,
|
| 480 |
+
"step": 300
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 0.03,
|
| 484 |
+
"eval_cos_sim": 0.8799148201942444,
|
| 485 |
+
"eval_loss": 0.12105432750927878,
|
| 486 |
+
"eval_runtime": 162.6443,
|
| 487 |
+
"eval_samples_per_second": 24.594,
|
| 488 |
+
"eval_steps_per_second": 0.387,
|
| 489 |
+
"step": 300
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 0.031,
|
| 493 |
+
"grad_norm": 0.027923179790377617,
|
| 494 |
+
"learning_rate": 4.947535513144286e-05,
|
| 495 |
+
"loss": 0.1197,
|
| 496 |
+
"step": 310
|
| 497 |
+
},
|
| 498 |
+
{
|
| 499 |
+
"epoch": 0.031,
|
| 500 |
+
"eval_cos_sim": 0.8799843788146973,
|
| 501 |
+
"eval_loss": 0.120985016367311,
|
| 502 |
+
"eval_runtime": 165.6185,
|
| 503 |
+
"eval_samples_per_second": 24.152,
|
| 504 |
+
"eval_steps_per_second": 0.38,
|
| 505 |
+
"step": 310
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 0.032,
|
| 509 |
+
"grad_norm": 0.025140805169939995,
|
| 510 |
+
"learning_rate": 1.9712165353304617e-05,
|
| 511 |
+
"loss": 0.1199,
|
| 512 |
+
"step": 320
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 0.032,
|
| 516 |
+
"eval_cos_sim": 0.8800230622291565,
|
| 517 |
+
"eval_loss": 0.1209463920806594,
|
| 518 |
+
"eval_runtime": 161.2564,
|
| 519 |
+
"eval_samples_per_second": 24.805,
|
| 520 |
+
"eval_steps_per_second": 0.391,
|
| 521 |
+
"step": 320
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.033,
|
| 525 |
+
"grad_norm": 0.03448393940925598,
|
| 526 |
+
"learning_rate": 4.9391877493394335e-05,
|
| 527 |
+
"loss": 0.1205,
|
| 528 |
+
"step": 330
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.033,
|
| 532 |
+
"eval_cos_sim": 0.8800433278083801,
|
| 533 |
+
"eval_loss": 0.12092636968838645,
|
| 534 |
+
"eval_runtime": 165.583,
|
| 535 |
+
"eval_samples_per_second": 24.157,
|
| 536 |
+
"eval_steps_per_second": 0.38,
|
| 537 |
+
"step": 330
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"epoch": 0.034,
|
| 541 |
+
"grad_norm": 0.027445893734693527,
|
| 542 |
+
"learning_rate": 1.9327097275960212e-05,
|
| 543 |
+
"loss": 0.1208,
|
| 544 |
+
"step": 340
|
| 545 |
+
},
|
| 546 |
+
{
|
| 547 |
+
"epoch": 0.034,
|
| 548 |
+
"eval_cos_sim": 0.8797867894172668,
|
| 549 |
+
"eval_loss": 0.12118107464062644,
|
| 550 |
+
"eval_runtime": 169.9936,
|
| 551 |
+
"eval_samples_per_second": 23.53,
|
| 552 |
+
"eval_steps_per_second": 0.371,
|
| 553 |
+
"step": 340
|
| 554 |
+
},
|
| 555 |
+
{
|
| 556 |
+
"epoch": 0.035,
|
| 557 |
+
"grad_norm": 0.032430149614810944,
|
| 558 |
+
"learning_rate": 4.9302320888106454e-05,
|
| 559 |
+
"loss": 0.1192,
|
| 560 |
+
"step": 350
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 0.035,
|
| 564 |
+
"eval_cos_sim": 0.8799253106117249,
|
| 565 |
+
"eval_loss": 0.12104199745404196,
|
| 566 |
+
"eval_runtime": 162.0388,
|
| 567 |
+
"eval_samples_per_second": 24.685,
|
| 568 |
+
"eval_steps_per_second": 0.389,
|
| 569 |
+
"step": 350
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.036,
|
| 573 |
+
"grad_norm": 0.03066575713455677,
|
| 574 |
+
"learning_rate": 1.894344300490539e-05,
|
| 575 |
+
"loss": 0.1207,
|
| 576 |
+
"step": 360
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.036,
|
| 580 |
+
"eval_cos_sim": 0.8800117373466492,
|
| 581 |
+
"eval_loss": 0.12095709469067527,
|
| 582 |
+
"eval_runtime": 168.2521,
|
| 583 |
+
"eval_samples_per_second": 23.774,
|
| 584 |
+
"eval_steps_per_second": 0.374,
|
| 585 |
+
"step": 360
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 0.037,
|
| 589 |
+
"grad_norm": 0.04023744910955429,
|
| 590 |
+
"learning_rate": 4.920670763496264e-05,
|
| 591 |
+
"loss": 0.1206,
|
| 592 |
+
"step": 370
|
| 593 |
+
},
|
| 594 |
+
{
|
| 595 |
+
"epoch": 0.037,
|
| 596 |
+
"eval_cos_sim": 0.8800341486930847,
|
| 597 |
+
"eval_loss": 0.12093350794064474,
|
| 598 |
+
"eval_runtime": 164.2446,
|
| 599 |
+
"eval_samples_per_second": 24.354,
|
| 600 |
+
"eval_steps_per_second": 0.384,
|
| 601 |
+
"step": 370
|
| 602 |
+
},
|
| 603 |
+
{
|
| 604 |
+
"epoch": 0.038,
|
| 605 |
+
"grad_norm": 0.03345053270459175,
|
| 606 |
+
"learning_rate": 1.8561298154827563e-05,
|
| 607 |
+
"loss": 0.1207,
|
| 608 |
+
"step": 380
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 0.038,
|
| 612 |
+
"eval_cos_sim": 0.8800336122512817,
|
| 613 |
+
"eval_loss": 0.12093256904828024,
|
| 614 |
+
"eval_runtime": 158.5429,
|
| 615 |
+
"eval_samples_per_second": 25.23,
|
| 616 |
+
"eval_steps_per_second": 0.397,
|
| 617 |
+
"step": 380
|
| 618 |
+
},
|
| 619 |
+
{
|
| 620 |
+
"epoch": 0.039,
|
| 621 |
+
"grad_norm": 0.02383916825056076,
|
| 622 |
+
"learning_rate": 4.910506156279026e-05,
|
| 623 |
+
"loss": 0.1213,
|
| 624 |
+
"step": 390
|
| 625 |
+
},
|
| 626 |
+
{
|
| 627 |
+
"epoch": 0.039,
|
| 628 |
+
"eval_cos_sim": 0.8800181150436401,
|
| 629 |
+
"eval_loss": 0.12094816543805074,
|
| 630 |
+
"eval_runtime": 164.6828,
|
| 631 |
+
"eval_samples_per_second": 24.289,
|
| 632 |
+
"eval_steps_per_second": 0.383,
|
| 633 |
+
"step": 390
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.04,
|
| 637 |
+
"grad_norm": 0.03217790648341179,
|
| 638 |
+
"learning_rate": 1.8180757964234907e-05,
|
| 639 |
+
"loss": 0.1213,
|
| 640 |
+
"step": 400
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.04,
|
| 644 |
+
"eval_cos_sim": 0.8800681829452515,
|
| 645 |
+
"eval_loss": 0.12089718677746726,
|
| 646 |
+
"eval_runtime": 162.9873,
|
| 647 |
+
"eval_samples_per_second": 24.542,
|
| 648 |
+
"eval_steps_per_second": 0.387,
|
| 649 |
+
"step": 400
|
| 650 |
+
},
|
| 651 |
+
{
|
| 652 |
+
"epoch": 0.041,
|
| 653 |
+
"grad_norm": 0.03514571115374565,
|
| 654 |
+
"learning_rate": 4.8997408003921466e-05,
|
| 655 |
+
"loss": 0.1208,
|
| 656 |
+
"step": 410
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"epoch": 0.041,
|
| 660 |
+
"eval_cos_sim": 0.8801241517066956,
|
| 661 |
+
"eval_loss": 0.12084018089520407,
|
| 662 |
+
"eval_runtime": 167.298,
|
| 663 |
+
"eval_samples_per_second": 23.909,
|
| 664 |
+
"eval_steps_per_second": 0.377,
|
| 665 |
+
"step": 410
|
| 666 |
+
},
|
| 667 |
+
{
|
| 668 |
+
"epoch": 0.042,
|
| 669 |
+
"grad_norm": 0.03063860908150673,
|
| 670 |
+
"learning_rate": 1.780191727172083e-05,
|
| 671 |
+
"loss": 0.1207,
|
| 672 |
+
"step": 420
|
| 673 |
+
},
|
| 674 |
+
{
|
| 675 |
+
"epoch": 0.042,
|
| 676 |
+
"eval_cos_sim": 0.8799417018890381,
|
| 677 |
+
"eval_loss": 0.12102477314221335,
|
| 678 |
+
"eval_runtime": 165.8735,
|
| 679 |
+
"eval_samples_per_second": 24.115,
|
| 680 |
+
"eval_steps_per_second": 0.38,
|
| 681 |
+
"step": 420
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.043,
|
| 685 |
+
"grad_norm": 0.0319770872592926,
|
| 686 |
+
"learning_rate": 4.8883773787879826e-05,
|
| 687 |
+
"loss": 0.1205,
|
| 688 |
+
"step": 430
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.043,
|
| 692 |
+
"eval_cos_sim": 0.8800466060638428,
|
| 693 |
+
"eval_loss": 0.12091960479962302,
|
| 694 |
+
"eval_runtime": 163.3913,
|
| 695 |
+
"eval_samples_per_second": 24.481,
|
| 696 |
+
"eval_steps_per_second": 0.386,
|
| 697 |
+
"step": 430
|
| 698 |
+
},
|
| 699 |
+
{
|
| 700 |
+
"epoch": 0.044,
|
| 701 |
+
"grad_norm": 0.02543482929468155,
|
| 702 |
+
"learning_rate": 1.742487049232818e-05,
|
| 703 |
+
"loss": 0.1202,
|
| 704 |
+
"step": 440
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 0.044,
|
| 708 |
+
"eval_cos_sim": 0.8802583813667297,
|
| 709 |
+
"eval_loss": 0.12070597412335349,
|
| 710 |
+
"eval_runtime": 160.473,
|
| 711 |
+
"eval_samples_per_second": 24.926,
|
| 712 |
+
"eval_steps_per_second": 0.393,
|
| 713 |
+
"step": 440
|
| 714 |
+
},
|
| 715 |
+
{
|
| 716 |
+
"epoch": 0.045,
|
| 717 |
+
"grad_norm": 0.024107394739985466,
|
| 718 |
+
"learning_rate": 4.876418723469453e-05,
|
| 719 |
+
"loss": 0.1196,
|
| 720 |
+
"step": 450
|
| 721 |
+
},
|
| 722 |
+
{
|
| 723 |
+
"epoch": 0.045,
|
| 724 |
+
"eval_cos_sim": 0.8802623748779297,
|
| 725 |
+
"eval_loss": 0.12070202877270651,
|
| 726 |
+
"eval_runtime": 168.2567,
|
| 727 |
+
"eval_samples_per_second": 23.773,
|
| 728 |
+
"eval_steps_per_second": 0.374,
|
| 729 |
+
"step": 450
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 0.046,
|
| 733 |
+
"grad_norm": 0.04505016654729843,
|
| 734 |
+
"learning_rate": 1.7049711594019046e-05,
|
| 735 |
+
"loss": 0.1197,
|
| 736 |
+
"step": 460
|
| 737 |
+
},
|
| 738 |
+
{
|
| 739 |
+
"epoch": 0.046,
|
| 740 |
+
"eval_cos_sim": 0.8800992965698242,
|
| 741 |
+
"eval_loss": 0.12086664869534446,
|
| 742 |
+
"eval_runtime": 168.3415,
|
| 743 |
+
"eval_samples_per_second": 23.761,
|
| 744 |
+
"eval_steps_per_second": 0.374,
|
| 745 |
+
"step": 460
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 0.047,
|
| 749 |
+
"grad_norm": 0.026298915967345238,
|
| 750 |
+
"learning_rate": 4.8638678147841726e-05,
|
| 751 |
+
"loss": 0.1207,
|
| 752 |
+
"step": 470
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 0.047,
|
| 756 |
+
"eval_cos_sim": 0.8801872134208679,
|
| 757 |
+
"eval_loss": 0.12077882673489523,
|
| 758 |
+
"eval_runtime": 171.1708,
|
| 759 |
+
"eval_samples_per_second": 23.368,
|
| 760 |
+
"eval_steps_per_second": 0.368,
|
| 761 |
+
"step": 470
|
| 762 |
+
},
|
| 763 |
+
{
|
| 764 |
+
"epoch": 0.048,
|
| 765 |
+
"grad_norm": 0.04072026535868645,
|
| 766 |
+
"learning_rate": 1.667653407425599e-05,
|
| 767 |
+
"loss": 0.12,
|
| 768 |
+
"step": 480
|
| 769 |
+
},
|
| 770 |
+
{
|
| 771 |
+
"epoch": 0.048,
|
| 772 |
+
"eval_cos_sim": 0.8804126977920532,
|
| 773 |
+
"eval_loss": 0.12055267622219992,
|
| 774 |
+
"eval_runtime": 160.8906,
|
| 775 |
+
"eval_samples_per_second": 24.862,
|
| 776 |
+
"eval_steps_per_second": 0.392,
|
| 777 |
+
"step": 480
|
| 778 |
+
},
|
| 779 |
+
{
|
| 780 |
+
"epoch": 0.049,
|
| 781 |
+
"grad_norm": 0.02353891357779503,
|
| 782 |
+
"learning_rate": 4.850727780681685e-05,
|
| 783 |
+
"loss": 0.121,
|
| 784 |
+
"step": 490
|
| 785 |
+
},
|
| 786 |
+
{
|
| 787 |
+
"epoch": 0.049,
|
| 788 |
+
"eval_cos_sim": 0.8802867531776428,
|
| 789 |
+
"eval_loss": 0.12067857982861471,
|
| 790 |
+
"eval_runtime": 162.0814,
|
| 791 |
+
"eval_samples_per_second": 24.679,
|
| 792 |
+
"eval_steps_per_second": 0.389,
|
| 793 |
+
"step": 490
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.05,
|
| 797 |
+
"grad_norm": 0.03163010999560356,
|
| 798 |
+
"learning_rate": 1.6305430936700462e-05,
|
| 799 |
+
"loss": 0.1206,
|
| 800 |
+
"step": 500
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.05,
|
| 804 |
+
"eval_cos_sim": 0.8799078464508057,
|
| 805 |
+
"eval_loss": 0.12105564882504416,
|
| 806 |
+
"eval_runtime": 159.7041,
|
| 807 |
+
"eval_samples_per_second": 25.046,
|
| 808 |
+
"eval_steps_per_second": 0.394,
|
| 809 |
+
"step": 500
|
| 810 |
+
},
|
| 811 |
+
{
|
| 812 |
+
"epoch": 0.051,
|
| 813 |
+
"grad_norm": 0.03480914607644081,
|
| 814 |
+
"learning_rate": 4.8370018959339916e-05,
|
| 815 |
+
"loss": 0.1193,
|
| 816 |
+
"step": 510
|
| 817 |
+
},
|
| 818 |
+
{
|
| 819 |
+
"epoch": 0.051,
|
| 820 |
+
"eval_cos_sim": 0.8801398873329163,
|
| 821 |
+
"eval_loss": 0.12082168438183737,
|
| 822 |
+
"eval_runtime": 168.1565,
|
| 823 |
+
"eval_samples_per_second": 23.787,
|
| 824 |
+
"eval_steps_per_second": 0.375,
|
| 825 |
+
"step": 510
|
| 826 |
+
},
|
| 827 |
+
{
|
| 828 |
+
"epoch": 0.052,
|
| 829 |
+
"grad_norm": 0.031566403806209564,
|
| 830 |
+
"learning_rate": 1.5936494668034417e-05,
|
| 831 |
+
"loss": 0.1207,
|
| 832 |
+
"step": 520
|
| 833 |
+
},
|
| 834 |
+
{
|
| 835 |
+
"epoch": 0.052,
|
| 836 |
+
"eval_cos_sim": 0.8804723024368286,
|
| 837 |
+
"eval_loss": 0.12048741771923971,
|
| 838 |
+
"eval_runtime": 162.0779,
|
| 839 |
+
"eval_samples_per_second": 24.679,
|
| 840 |
+
"eval_steps_per_second": 0.389,
|
| 841 |
+
"step": 520
|
| 842 |
+
},
|
| 843 |
+
{
|
| 844 |
+
"epoch": 0.053,
|
| 845 |
+
"grad_norm": 0.02134857140481472,
|
| 846 |
+
"learning_rate": 4.822693581319333e-05,
|
| 847 |
+
"loss": 0.1207,
|
| 848 |
+
"step": 530
|
| 849 |
+
},
|
| 850 |
+
{
|
| 851 |
+
"epoch": 0.053,
|
| 852 |
+
"eval_cos_sim": 0.8804084062576294,
|
| 853 |
+
"eval_loss": 0.12055278637158347,
|
| 854 |
+
"eval_runtime": 165.0027,
|
| 855 |
+
"eval_samples_per_second": 24.242,
|
| 856 |
+
"eval_steps_per_second": 0.382,
|
| 857 |
+
"step": 530
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 0.054,
|
| 861 |
+
"grad_norm": 0.02998766116797924,
|
| 862 |
+
"learning_rate": 1.5569817214910634e-05,
|
| 863 |
+
"loss": 0.1206,
|
| 864 |
+
"step": 540
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 0.054,
|
| 868 |
+
"eval_cos_sim": 0.879996657371521,
|
| 869 |
+
"eval_loss": 0.12096652509915305,
|
| 870 |
+
"eval_runtime": 269.6523,
|
| 871 |
+
"eval_samples_per_second": 14.834,
|
| 872 |
+
"eval_steps_per_second": 0.234,
|
| 873 |
+
"step": 540
|
| 874 |
+
},
|
| 875 |
+
{
|
| 876 |
+
"epoch": 0.055,
|
| 877 |
+
"grad_norm": 0.023394938558340073,
|
| 878 |
+
"learning_rate": 4.807806402769648e-05,
|
| 879 |
+
"loss": 0.1204,
|
| 880 |
+
"step": 550
|
| 881 |
+
},
|
| 882 |
+
{
|
| 883 |
+
"epoch": 0.055,
|
| 884 |
+
"eval_cos_sim": 0.8802942037582397,
|
| 885 |
+
"eval_loss": 0.12066655971753074,
|
| 886 |
+
"eval_runtime": 241.2043,
|
| 887 |
+
"eval_samples_per_second": 16.583,
|
| 888 |
+
"eval_steps_per_second": 0.261,
|
| 889 |
+
"step": 550
|
| 890 |
+
},
|
| 891 |
+
{
|
| 892 |
+
"epoch": 0.056,
|
| 893 |
+
"grad_norm": 0.04035342484712601,
|
| 894 |
+
"learning_rate": 1.520548996103771e-05,
|
| 895 |
+
"loss": 0.1208,
|
| 896 |
+
"step": 560
|
| 897 |
+
},
|
| 898 |
+
{
|
| 899 |
+
"epoch": 0.056,
|
| 900 |
+
"eval_cos_sim": 0.8805367946624756,
|
| 901 |
+
"eval_loss": 0.12042092802273703,
|
| 902 |
+
"eval_runtime": 217.8859,
|
| 903 |
+
"eval_samples_per_second": 18.358,
|
| 904 |
+
"eval_steps_per_second": 0.289,
|
| 905 |
+
"step": 560
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.057,
|
| 909 |
+
"grad_norm": 0.02704194188117981,
|
| 910 |
+
"learning_rate": 4.7923440704819685e-05,
|
| 911 |
+
"loss": 0.1205,
|
| 912 |
+
"step": 570
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.057,
|
| 916 |
+
"eval_cos_sim": 0.8805016875267029,
|
| 917 |
+
"eval_loss": 0.12045616819607688,
|
| 918 |
+
"eval_runtime": 163.5639,
|
| 919 |
+
"eval_samples_per_second": 24.455,
|
| 920 |
+
"eval_steps_per_second": 0.385,
|
| 921 |
+
"step": 570
|
| 922 |
+
},
|
| 923 |
+
{
|
| 924 |
+
"epoch": 0.058,
|
| 925 |
+
"grad_norm": 0.041525471955537796,
|
| 926 |
+
"learning_rate": 1.4843603704405321e-05,
|
| 927 |
+
"loss": 0.1209,
|
| 928 |
+
"step": 580
|
| 929 |
+
},
|
| 930 |
+
{
|
| 931 |
+
"epoch": 0.058,
|
| 932 |
+
"eval_cos_sim": 0.8803950548171997,
|
| 933 |
+
"eval_loss": 0.12056205751645041,
|
| 934 |
+
"eval_runtime": 163.8454,
|
| 935 |
+
"eval_samples_per_second": 24.413,
|
| 936 |
+
"eval_steps_per_second": 0.385,
|
| 937 |
+
"step": 580
|
| 938 |
+
},
|
| 939 |
+
{
|
| 940 |
+
"epoch": 0.059,
|
| 941 |
+
"grad_norm": 0.02588295191526413,
|
| 942 |
+
"learning_rate": 4.7763104379936636e-05,
|
| 943 |
+
"loss": 0.12,
|
| 944 |
+
"step": 590
|
| 945 |
+
},
|
| 946 |
+
{
|
| 947 |
+
"epoch": 0.059,
|
| 948 |
+
"eval_cos_sim": 0.8804982304573059,
|
| 949 |
+
"eval_loss": 0.12045794346081687,
|
| 950 |
+
"eval_runtime": 175.2999,
|
| 951 |
+
"eval_samples_per_second": 22.818,
|
| 952 |
+
"eval_steps_per_second": 0.359,
|
| 953 |
+
"step": 590
|
| 954 |
+
},
|
| 955 |
+
{
|
| 956 |
+
"epoch": 0.06,
|
| 957 |
+
"grad_norm": 0.030644405633211136,
|
| 958 |
+
"learning_rate": 1.4484248634655188e-05,
|
| 959 |
+
"loss": 0.1211,
|
| 960 |
+
"step": 600
|
| 961 |
+
},
|
| 962 |
+
{
|
| 963 |
+
"epoch": 0.06,
|
| 964 |
+
"eval_cos_sim": 0.8804518580436707,
|
| 965 |
+
"eval_loss": 0.12050184681164694,
|
| 966 |
+
"eval_runtime": 165.1748,
|
| 967 |
+
"eval_samples_per_second": 24.217,
|
| 968 |
+
"eval_steps_per_second": 0.381,
|
| 969 |
+
"step": 600
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 0.061,
|
| 973 |
+
"grad_norm": 0.03162102401256561,
|
| 974 |
+
"learning_rate": 4.7597095012220556e-05,
|
| 975 |
+
"loss": 0.1194,
|
| 976 |
+
"step": 610
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 0.061,
|
| 980 |
+
"eval_cos_sim": 0.8805278539657593,
|
| 981 |
+
"eval_loss": 0.12042546226727438,
|
| 982 |
+
"eval_runtime": 169.0168,
|
| 983 |
+
"eval_samples_per_second": 23.666,
|
| 984 |
+
"eval_steps_per_second": 0.373,
|
| 985 |
+
"step": 610
|
| 986 |
+
},
|
| 987 |
+
{
|
| 988 |
+
"epoch": 0.062,
|
| 989 |
+
"grad_norm": 0.030952898785471916,
|
| 990 |
+
"learning_rate": 1.4127514310605238e-05,
|
| 991 |
+
"loss": 0.1202,
|
| 992 |
+
"step": 620
|
| 993 |
+
},
|
| 994 |
+
{
|
| 995 |
+
"epoch": 0.062,
|
| 996 |
+
"eval_cos_sim": 0.880526602268219,
|
| 997 |
+
"eval_loss": 0.12043014385449362,
|
| 998 |
+
"eval_runtime": 161.73,
|
| 999 |
+
"eval_samples_per_second": 24.733,
|
| 1000 |
+
"eval_steps_per_second": 0.39,
|
| 1001 |
+
"step": 620
|
| 1002 |
+
},
|
| 1003 |
+
{
|
| 1004 |
+
"epoch": 0.063,
|
| 1005 |
+
"grad_norm": 0.025900695472955704,
|
| 1006 |
+
"learning_rate": 4.742545397468656e-05,
|
| 1007 |
+
"loss": 0.1205,
|
| 1008 |
+
"step": 630
|
| 1009 |
+
},
|
| 1010 |
+
{
|
| 1011 |
+
"epoch": 0.063,
|
| 1012 |
+
"eval_cos_sim": 0.8804138898849487,
|
| 1013 |
+
"eval_loss": 0.12054368068921043,
|
| 1014 |
+
"eval_runtime": 169.1165,
|
| 1015 |
+
"eval_samples_per_second": 23.652,
|
| 1016 |
+
"eval_steps_per_second": 0.373,
|
| 1017 |
+
"step": 630
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.064,
|
| 1021 |
+
"grad_norm": 0.02121679112315178,
|
| 1022 |
+
"learning_rate": 1.3773489637927061e-05,
|
| 1023 |
+
"loss": 0.1208,
|
| 1024 |
+
"step": 640
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 0.064,
|
| 1028 |
+
"eval_cos_sim": 0.8801329731941223,
|
| 1029 |
+
"eval_loss": 0.120824953577394,
|
| 1030 |
+
"eval_runtime": 161.444,
|
| 1031 |
+
"eval_samples_per_second": 24.776,
|
| 1032 |
+
"eval_steps_per_second": 0.39,
|
| 1033 |
+
"step": 640
|
| 1034 |
+
},
|
| 1035 |
+
{
|
| 1036 |
+
"epoch": 0.065,
|
| 1037 |
+
"grad_norm": 0.02153482660651207,
|
| 1038 |
+
"learning_rate": 4.7248224043879605e-05,
|
| 1039 |
+
"loss": 0.1211,
|
| 1040 |
+
"step": 650
|
| 1041 |
+
},
|
| 1042 |
+
{
|
| 1043 |
+
"epoch": 0.065,
|
| 1044 |
+
"eval_cos_sim": 0.8802359104156494,
|
| 1045 |
+
"eval_loss": 0.12072229530560447,
|
| 1046 |
+
"eval_runtime": 161.7761,
|
| 1047 |
+
"eval_samples_per_second": 24.726,
|
| 1048 |
+
"eval_steps_per_second": 0.389,
|
| 1049 |
+
"step": 650
|
| 1050 |
+
},
|
| 1051 |
+
{
|
| 1052 |
+
"epoch": 0.066,
|
| 1053 |
+
"grad_norm": 0.030305592343211174,
|
| 1054 |
+
"learning_rate": 1.342226284699112e-05,
|
| 1055 |
+
"loss": 0.1202,
|
| 1056 |
+
"step": 660
|
| 1057 |
+
},
|
| 1058 |
+
{
|
| 1059 |
+
"epoch": 0.066,
|
| 1060 |
+
"eval_cos_sim": 0.8804138898849487,
|
| 1061 |
+
"eval_loss": 0.12054470445858909,
|
| 1062 |
+
"eval_runtime": 166.8157,
|
| 1063 |
+
"eval_samples_per_second": 23.979,
|
| 1064 |
+
"eval_steps_per_second": 0.378,
|
| 1065 |
+
"step": 660
|
| 1066 |
+
},
|
| 1067 |
+
{
|
| 1068 |
+
"epoch": 0.067,
|
| 1069 |
+
"grad_norm": 0.03320358693599701,
|
| 1070 |
+
"learning_rate": 4.7065449389213644e-05,
|
| 1071 |
+
"loss": 0.1216,
|
| 1072 |
+
"step": 670
|
| 1073 |
+
},
|
| 1074 |
+
{
|
| 1075 |
+
"epoch": 0.067,
|
| 1076 |
+
"eval_cos_sim": 0.8803730607032776,
|
| 1077 |
+
"eval_loss": 0.12058561565625144,
|
| 1078 |
+
"eval_runtime": 168.9555,
|
| 1079 |
+
"eval_samples_per_second": 23.675,
|
| 1080 |
+
"eval_steps_per_second": 0.373,
|
| 1081 |
+
"step": 670
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 0.068,
|
| 1085 |
+
"grad_norm": 0.03419356420636177,
|
| 1086 |
+
"learning_rate": 1.3073921470878081e-05,
|
| 1087 |
+
"loss": 0.1197,
|
| 1088 |
+
"step": 680
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 0.068,
|
| 1092 |
+
"eval_cos_sim": 0.8803737163543701,
|
| 1093 |
+
"eval_loss": 0.12058660843121481,
|
| 1094 |
+
"eval_runtime": 162.073,
|
| 1095 |
+
"eval_samples_per_second": 24.68,
|
| 1096 |
+
"eval_steps_per_second": 0.389,
|
| 1097 |
+
"step": 680
|
| 1098 |
+
},
|
| 1099 |
+
{
|
| 1100 |
+
"epoch": 0.069,
|
| 1101 |
+
"grad_norm": 0.022392097860574722,
|
| 1102 |
+
"learning_rate": 4.6877175561964684e-05,
|
| 1103 |
+
"loss": 0.12,
|
| 1104 |
+
"step": 690
|
| 1105 |
+
},
|
| 1106 |
+
{
|
| 1107 |
+
"epoch": 0.069,
|
| 1108 |
+
"eval_cos_sim": 0.8804360628128052,
|
| 1109 |
+
"eval_loss": 0.12052560474621725,
|
| 1110 |
+
"eval_runtime": 184.9748,
|
| 1111 |
+
"eval_samples_per_second": 21.625,
|
| 1112 |
+
"eval_steps_per_second": 0.341,
|
| 1113 |
+
"step": 690
|
| 1114 |
+
},
|
| 1115 |
+
{
|
| 1116 |
+
"epoch": 0.07,
|
| 1117 |
+
"grad_norm": 0.03007390908896923,
|
| 1118 |
+
"learning_rate": 1.272855232356e-05,
|
| 1119 |
+
"loss": 0.1204,
|
| 1120 |
+
"step": 700
|
| 1121 |
+
},
|
| 1122 |
+
{
|
| 1123 |
+
"epoch": 0.07,
|
| 1124 |
+
"eval_cos_sim": 0.8805059790611267,
|
| 1125 |
+
"eval_loss": 0.12045389034497214,
|
| 1126 |
+
"eval_runtime": 165.5078,
|
| 1127 |
+
"eval_samples_per_second": 24.168,
|
| 1128 |
+
"eval_steps_per_second": 0.381,
|
| 1129 |
+
"step": 700
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 0.071,
|
| 1133 |
+
"grad_norm": 0.020742209628224373,
|
| 1134 |
+
"learning_rate": 4.6683449483917846e-05,
|
| 1135 |
+
"loss": 0.12,
|
| 1136 |
+
"step": 710
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 0.071,
|
| 1140 |
+
"eval_cos_sim": 0.8806586861610413,
|
| 1141 |
+
"eval_loss": 0.12030088522183371,
|
| 1142 |
+
"eval_runtime": 172.1855,
|
| 1143 |
+
"eval_samples_per_second": 23.231,
|
| 1144 |
+
"eval_steps_per_second": 0.366,
|
| 1145 |
+
"step": 710
|
| 1146 |
+
},
|
| 1147 |
+
{
|
| 1148 |
+
"epoch": 0.072,
|
| 1149 |
+
"grad_norm": 0.023063719272613525,
|
| 1150 |
+
"learning_rate": 1.2386241478270652e-05,
|
| 1151 |
+
"loss": 0.1198,
|
| 1152 |
+
"step": 720
|
| 1153 |
+
},
|
| 1154 |
+
{
|
| 1155 |
+
"epoch": 0.072,
|
| 1156 |
+
"eval_cos_sim": 0.8805540204048157,
|
| 1157 |
+
"eval_loss": 0.12040612079846334,
|
| 1158 |
+
"eval_runtime": 166.7801,
|
| 1159 |
+
"eval_samples_per_second": 23.984,
|
| 1160 |
+
"eval_steps_per_second": 0.378,
|
| 1161 |
+
"step": 720
|
| 1162 |
+
},
|
| 1163 |
+
{
|
| 1164 |
+
"epoch": 0.073,
|
| 1165 |
+
"grad_norm": 0.027647124603390694,
|
| 1166 |
+
"learning_rate": 4.648431943567264e-05,
|
| 1167 |
+
"loss": 0.1205,
|
| 1168 |
+
"step": 730
|
| 1169 |
+
},
|
| 1170 |
+
{
|
| 1171 |
+
"epoch": 0.073,
|
| 1172 |
+
"eval_cos_sim": 0.8805664777755737,
|
| 1173 |
+
"eval_loss": 0.12039038088070822,
|
| 1174 |
+
"eval_runtime": 164.5025,
|
| 1175 |
+
"eval_samples_per_second": 24.316,
|
| 1176 |
+
"eval_steps_per_second": 0.383,
|
| 1177 |
+
"step": 730
|
| 1178 |
+
},
|
| 1179 |
+
{
|
| 1180 |
+
"epoch": 0.074,
|
| 1181 |
+
"grad_norm": 0.02208826318383217,
|
| 1182 |
+
"learning_rate": 1.204707424604792e-05,
|
| 1183 |
+
"loss": 0.1203,
|
| 1184 |
+
"step": 740
|
| 1185 |
+
},
|
| 1186 |
+
{
|
| 1187 |
+
"epoch": 0.074,
|
| 1188 |
+
"eval_cos_sim": 0.8805152177810669,
|
| 1189 |
+
"eval_loss": 0.12043928005444479,
|
| 1190 |
+
"eval_runtime": 178.0321,
|
| 1191 |
+
"eval_samples_per_second": 22.468,
|
| 1192 |
+
"eval_steps_per_second": 0.354,
|
| 1193 |
+
"step": 740
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 0.075,
|
| 1197 |
+
"grad_norm": 0.021549325436353683,
|
| 1198 |
+
"learning_rate": 4.627983504461235e-05,
|
| 1199 |
+
"loss": 0.1196,
|
| 1200 |
+
"step": 750
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 0.075,
|
| 1204 |
+
"eval_cos_sim": 0.8806087374687195,
|
| 1205 |
+
"eval_loss": 0.12034820940243673,
|
| 1206 |
+
"eval_runtime": 179.3479,
|
| 1207 |
+
"eval_samples_per_second": 22.303,
|
| 1208 |
+
"eval_steps_per_second": 0.351,
|
| 1209 |
+
"step": 750
|
| 1210 |
+
},
|
| 1211 |
+
{
|
| 1212 |
+
"epoch": 0.076,
|
| 1213 |
+
"grad_norm": 0.028022369369864464,
|
| 1214 |
+
"learning_rate": 1.1711135154477562e-05,
|
| 1215 |
+
"loss": 0.1207,
|
| 1216 |
+
"step": 760
|
| 1217 |
+
},
|
| 1218 |
+
{
|
| 1219 |
+
"epoch": 0.076,
|
| 1220 |
+
"eval_cos_sim": 0.8805749416351318,
|
| 1221 |
+
"eval_loss": 0.12038306286084127,
|
| 1222 |
+
"eval_runtime": 176.8723,
|
| 1223 |
+
"eval_samples_per_second": 22.615,
|
| 1224 |
+
"eval_steps_per_second": 0.356,
|
| 1225 |
+
"step": 760
|
| 1226 |
+
},
|
| 1227 |
+
{
|
| 1228 |
+
"epoch": 0.077,
|
| 1229 |
+
"grad_norm": 0.021017303690314293,
|
| 1230 |
+
"learning_rate": 4.607004727253391e-05,
|
| 1231 |
+
"loss": 0.12,
|
| 1232 |
+
"step": 770
|
| 1233 |
+
},
|
| 1234 |
+
{
|
| 1235 |
+
"epoch": 0.077,
|
| 1236 |
+
"eval_cos_sim": 0.8806374669075012,
|
| 1237 |
+
"eval_loss": 0.12032015850293112,
|
| 1238 |
+
"eval_runtime": 163.1866,
|
| 1239 |
+
"eval_samples_per_second": 24.512,
|
| 1240 |
+
"eval_steps_per_second": 0.386,
|
| 1241 |
+
"step": 770
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 0.078,
|
| 1245 |
+
"grad_norm": 0.02246786840260029,
|
| 1246 |
+
"learning_rate": 1.1378507926623572e-05,
|
| 1247 |
+
"loss": 0.1199,
|
| 1248 |
+
"step": 780
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 0.078,
|
| 1252 |
+
"eval_cos_sim": 0.8806332349777222,
|
| 1253 |
+
"eval_loss": 0.12032350780713034,
|
| 1254 |
+
"eval_runtime": 163.0562,
|
| 1255 |
+
"eval_samples_per_second": 24.531,
|
| 1256 |
+
"eval_steps_per_second": 0.386,
|
| 1257 |
+
"step": 780
|
| 1258 |
+
},
|
| 1259 |
+
{
|
| 1260 |
+
"epoch": 0.079,
|
| 1261 |
+
"grad_norm": 0.021708086133003235,
|
| 1262 |
+
"learning_rate": 4.585500840294793e-05,
|
| 1263 |
+
"loss": 0.1201,
|
| 1264 |
+
"step": 790
|
| 1265 |
+
},
|
| 1266 |
+
{
|
| 1267 |
+
"epoch": 0.079,
|
| 1268 |
+
"eval_cos_sim": 0.8807349801063538,
|
| 1269 |
+
"eval_loss": 0.12022322513806295,
|
| 1270 |
+
"eval_runtime": 173.5217,
|
| 1271 |
+
"eval_samples_per_second": 23.052,
|
| 1272 |
+
"eval_steps_per_second": 0.363,
|
| 1273 |
+
"step": 790
|
| 1274 |
+
},
|
| 1275 |
+
{
|
| 1276 |
+
"epoch": 0.08,
|
| 1277 |
+
"grad_norm": 0.034823887050151825,
|
| 1278 |
+
"learning_rate": 1.1049275460164102e-05,
|
| 1279 |
+
"loss": 0.1204,
|
| 1280 |
+
"step": 800
|
| 1281 |
+
},
|
| 1282 |
+
{
|
| 1283 |
+
"epoch": 0.08,
|
| 1284 |
+
"eval_cos_sim": 0.8806157112121582,
|
| 1285 |
+
"eval_loss": 0.12034289600598289,
|
| 1286 |
+
"eval_runtime": 163.4003,
|
| 1287 |
+
"eval_samples_per_second": 24.48,
|
| 1288 |
+
"eval_steps_per_second": 0.386,
|
| 1289 |
+
"step": 800
|
| 1290 |
+
},
|
| 1291 |
+
{
|
| 1292 |
+
"epoch": 0.081,
|
| 1293 |
+
"grad_norm": 0.02099907584488392,
|
| 1294 |
+
"learning_rate": 4.563477202804924e-05,
|
| 1295 |
+
"loss": 0.1203,
|
| 1296 |
+
"step": 810
|
| 1297 |
+
},
|
| 1298 |
+
{
|
| 1299 |
+
"epoch": 0.081,
|
| 1300 |
+
"eval_cos_sim": 0.8805558681488037,
|
| 1301 |
+
"eval_loss": 0.1204009156440444,
|
| 1302 |
+
"eval_runtime": 171.4073,
|
| 1303 |
+
"eval_samples_per_second": 23.336,
|
| 1304 |
+
"eval_steps_per_second": 0.368,
|
| 1305 |
+
"step": 810
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 0.082,
|
| 1309 |
+
"grad_norm": 0.027718910947442055,
|
| 1310 |
+
"learning_rate": 1.0723519806732512e-05,
|
| 1311 |
+
"loss": 0.1206,
|
| 1312 |
+
"step": 820
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 0.082,
|
| 1316 |
+
"eval_cos_sim": 0.8804323673248291,
|
| 1317 |
+
"eval_loss": 0.12052262928235007,
|
| 1318 |
+
"eval_runtime": 166.5171,
|
| 1319 |
+
"eval_samples_per_second": 24.022,
|
| 1320 |
+
"eval_steps_per_second": 0.378,
|
| 1321 |
+
"step": 820
|
| 1322 |
+
},
|
| 1323 |
+
{
|
| 1324 |
+
"epoch": 0.083,
|
| 1325 |
+
"grad_norm": 0.030117569491267204,
|
| 1326 |
+
"learning_rate": 4.540939303535997e-05,
|
| 1327 |
+
"loss": 0.1208,
|
| 1328 |
+
"step": 830
|
| 1329 |
+
},
|
| 1330 |
+
{
|
| 1331 |
+
"epoch": 0.083,
|
| 1332 |
+
"eval_cos_sim": 0.8806024193763733,
|
| 1333 |
+
"eval_loss": 0.12035309459912252,
|
| 1334 |
+
"eval_runtime": 172.9809,
|
| 1335 |
+
"eval_samples_per_second": 23.124,
|
| 1336 |
+
"eval_steps_per_second": 0.364,
|
| 1337 |
+
"step": 830
|
| 1338 |
+
},
|
| 1339 |
+
{
|
| 1340 |
+
"epoch": 0.084,
|
| 1341 |
+
"grad_norm": 0.025621019303798676,
|
| 1342 |
+
"learning_rate": 1.0401322151467458e-05,
|
| 1343 |
+
"loss": 0.1207,
|
| 1344 |
+
"step": 840
|
| 1345 |
+
},
|
| 1346 |
+
{
|
| 1347 |
+
"epoch": 0.084,
|
| 1348 |
+
"eval_cos_sim": 0.8805838823318481,
|
| 1349 |
+
"eval_loss": 0.12037316419827414,
|
| 1350 |
+
"eval_runtime": 161.0298,
|
| 1351 |
+
"eval_samples_per_second": 24.84,
|
| 1352 |
+
"eval_steps_per_second": 0.391,
|
| 1353 |
+
"step": 840
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 0.085,
|
| 1357 |
+
"grad_norm": 0.037043701857328415,
|
| 1358 |
+
"learning_rate": 4.517892759404947e-05,
|
| 1359 |
+
"loss": 0.1192,
|
| 1360 |
+
"step": 850
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 0.085,
|
| 1364 |
+
"eval_cos_sim": 0.8807790279388428,
|
| 1365 |
+
"eval_loss": 0.12017762710797263,
|
| 1366 |
+
"eval_runtime": 165.404,
|
| 1367 |
+
"eval_samples_per_second": 24.183,
|
| 1368 |
+
"eval_steps_per_second": 0.381,
|
| 1369 |
+
"step": 850
|
| 1370 |
+
},
|
| 1371 |
+
{
|
| 1372 |
+
"epoch": 0.086,
|
| 1373 |
+
"grad_norm": 0.024647973477840424,
|
| 1374 |
+
"learning_rate": 1.0082762792778497e-05,
|
| 1375 |
+
"loss": 0.12,
|
| 1376 |
+
"step": 860
|
| 1377 |
+
},
|
| 1378 |
+
{
|
| 1379 |
+
"epoch": 0.086,
|
| 1380 |
+
"eval_cos_sim": 0.8808472156524658,
|
| 1381 |
+
"eval_loss": 0.12010916282879783,
|
| 1382 |
+
"eval_runtime": 159.4806,
|
| 1383 |
+
"eval_samples_per_second": 25.081,
|
| 1384 |
+
"eval_steps_per_second": 0.395,
|
| 1385 |
+
"step": 860
|
| 1386 |
+
},
|
| 1387 |
+
{
|
| 1388 |
+
"epoch": 0.087,
|
| 1389 |
+
"grad_norm": 0.02787039987742901,
|
| 1390 |
+
"learning_rate": 4.494343314093799e-05,
|
| 1391 |
+
"loss": 0.1192,
|
| 1392 |
+
"step": 870
|
| 1393 |
+
},
|
| 1394 |
+
{
|
| 1395 |
+
"epoch": 0.087,
|
| 1396 |
+
"eval_cos_sim": 0.8806514143943787,
|
| 1397 |
+
"eval_loss": 0.12030471136319112,
|
| 1398 |
+
"eval_runtime": 167.7185,
|
| 1399 |
+
"eval_samples_per_second": 23.849,
|
| 1400 |
+
"eval_steps_per_second": 0.376,
|
| 1401 |
+
"step": 870
|
| 1402 |
+
},
|
| 1403 |
+
{
|
| 1404 |
+
"epoch": 0.088,
|
| 1405 |
+
"grad_norm": 0.027198661118745804,
|
| 1406 |
+
"learning_rate": 9.767921122337203e-06,
|
| 1407 |
+
"loss": 0.12,
|
| 1408 |
+
"step": 880
|
| 1409 |
+
},
|
| 1410 |
+
{
|
| 1411 |
+
"epoch": 0.088,
|
| 1412 |
+
"eval_cos_sim": 0.8806140422821045,
|
| 1413 |
+
"eval_loss": 0.12034183456646871,
|
| 1414 |
+
"eval_runtime": 164.5125,
|
| 1415 |
+
"eval_samples_per_second": 24.314,
|
| 1416 |
+
"eval_steps_per_second": 0.383,
|
| 1417 |
+
"step": 880
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 0.089,
|
| 1421 |
+
"grad_norm": 0.020295780152082443,
|
| 1422 |
+
"learning_rate": 4.4702968366179995e-05,
|
| 1423 |
+
"loss": 0.121,
|
| 1424 |
+
"step": 890
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 0.089,
|
| 1428 |
+
"eval_cos_sim": 0.8807177543640137,
|
| 1429 |
+
"eval_loss": 0.12023961307751609,
|
| 1430 |
+
"eval_runtime": 178.9283,
|
| 1431 |
+
"eval_samples_per_second": 22.355,
|
| 1432 |
+
"eval_steps_per_second": 0.352,
|
| 1433 |
+
"step": 890
|
| 1434 |
+
},
|
| 1435 |
+
{
|
| 1436 |
+
"epoch": 0.09,
|
| 1437 |
+
"grad_norm": 0.025682412087917328,
|
| 1438 |
+
"learning_rate": 9.456875605287963e-06,
|
| 1439 |
+
"loss": 0.1197,
|
| 1440 |
+
"step": 900
|
| 1441 |
+
},
|
| 1442 |
+
{
|
| 1443 |
+
"epoch": 0.09,
|
| 1444 |
+
"eval_cos_sim": 0.8808146715164185,
|
| 1445 |
+
"eval_loss": 0.12014213421093893,
|
| 1446 |
+
"eval_runtime": 162.1503,
|
| 1447 |
+
"eval_samples_per_second": 24.668,
|
| 1448 |
+
"eval_steps_per_second": 0.389,
|
| 1449 |
+
"step": 900
|
| 1450 |
+
},
|
| 1451 |
+
{
|
| 1452 |
+
"epoch": 0.091,
|
| 1453 |
+
"grad_norm": 0.0241321362555027,
|
| 1454 |
+
"learning_rate": 4.4457593198638266e-05,
|
| 1455 |
+
"loss": 0.1204,
|
| 1456 |
+
"step": 910
|
| 1457 |
+
},
|
| 1458 |
+
{
|
| 1459 |
+
"epoch": 0.091,
|
| 1460 |
+
"eval_cos_sim": 0.8806710243225098,
|
| 1461 |
+
"eval_loss": 0.12028472664105368,
|
| 1462 |
+
"eval_runtime": 170.9864,
|
| 1463 |
+
"eval_samples_per_second": 23.394,
|
| 1464 |
+
"eval_steps_per_second": 0.368,
|
| 1465 |
+
"step": 910
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 0.092,
|
| 1469 |
+
"grad_norm": 0.03511843457818031,
|
| 1470 |
+
"learning_rate": 9.149703760693733e-06,
|
| 1471 |
+
"loss": 0.1204,
|
| 1472 |
+
"step": 920
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 0.092,
|
| 1476 |
+
"eval_cos_sim": 0.8806130886077881,
|
| 1477 |
+
"eval_loss": 0.12034211971508932,
|
| 1478 |
+
"eval_runtime": 163.6776,
|
| 1479 |
+
"eval_samples_per_second": 24.438,
|
| 1480 |
+
"eval_steps_per_second": 0.385,
|
| 1481 |
+
"step": 920
|
| 1482 |
+
},
|
| 1483 |
+
{
|
| 1484 |
+
"epoch": 0.093,
|
| 1485 |
+
"grad_norm": 0.03159726411104202,
|
| 1486 |
+
"learning_rate": 4.420736879094911e-05,
|
| 1487 |
+
"loss": 0.1208,
|
| 1488 |
+
"step": 930
|
| 1489 |
+
},
|
| 1490 |
+
{
|
| 1491 |
+
"epoch": 0.093,
|
| 1492 |
+
"eval_cos_sim": 0.8807392716407776,
|
| 1493 |
+
"eval_loss": 0.12021884395825339,
|
| 1494 |
+
"eval_runtime": 172.0497,
|
| 1495 |
+
"eval_samples_per_second": 23.249,
|
| 1496 |
+
"eval_steps_per_second": 0.366,
|
| 1497 |
+
"step": 930
|
| 1498 |
+
},
|
| 1499 |
+
{
|
| 1500 |
+
"epoch": 0.094,
|
| 1501 |
+
"grad_norm": 0.02288082055747509,
|
| 1502 |
+
"learning_rate": 8.846482142219678e-06,
|
| 1503 |
+
"loss": 0.1206,
|
| 1504 |
+
"step": 940
|
| 1505 |
+
},
|
| 1506 |
+
{
|
| 1507 |
+
"epoch": 0.094,
|
| 1508 |
+
"eval_cos_sim": 0.8808532953262329,
|
| 1509 |
+
"eval_loss": 0.12010584022747946,
|
| 1510 |
+
"eval_runtime": 160.6917,
|
| 1511 |
+
"eval_samples_per_second": 24.892,
|
| 1512 |
+
"eval_steps_per_second": 0.392,
|
| 1513 |
+
"step": 940
|
| 1514 |
+
},
|
| 1515 |
+
{
|
| 1516 |
+
"epoch": 0.095,
|
| 1517 |
+
"grad_norm": 0.022692304104566574,
|
| 1518 |
+
"learning_rate": 4.395235750428116e-05,
|
| 1519 |
+
"loss": 0.1193,
|
| 1520 |
+
"step": 950
|
| 1521 |
+
},
|
| 1522 |
+
{
|
| 1523 |
+
"epoch": 0.095,
|
| 1524 |
+
"eval_cos_sim": 0.8807929158210754,
|
| 1525 |
+
"eval_loss": 0.1201639943336196,
|
| 1526 |
+
"eval_runtime": 165.9011,
|
| 1527 |
+
"eval_samples_per_second": 24.111,
|
| 1528 |
+
"eval_steps_per_second": 0.38,
|
| 1529 |
+
"step": 950
|
| 1530 |
+
},
|
| 1531 |
+
{
|
| 1532 |
+
"epoch": 0.096,
|
| 1533 |
+
"grad_norm": 0.02069064788520336,
|
| 1534 |
+
"learning_rate": 8.547286319049193e-06,
|
| 1535 |
+
"loss": 0.1204,
|
| 1536 |
+
"step": 960
|
| 1537 |
+
},
|
| 1538 |
+
{
|
| 1539 |
+
"epoch": 0.096,
|
| 1540 |
+
"eval_cos_sim": 0.8806983232498169,
|
| 1541 |
+
"eval_loss": 0.12025791980969382,
|
| 1542 |
+
"eval_runtime": 163.2536,
|
| 1543 |
+
"eval_samples_per_second": 24.502,
|
| 1544 |
+
"eval_steps_per_second": 0.386,
|
| 1545 |
+
"step": 960
|
| 1546 |
+
},
|
| 1547 |
+
{
|
| 1548 |
+
"epoch": 0.097,
|
| 1549 |
+
"grad_norm": 0.024405937641859055,
|
| 1550 |
+
"learning_rate": 4.369262289279257e-05,
|
| 1551 |
+
"loss": 0.12,
|
| 1552 |
+
"step": 970
|
| 1553 |
+
},
|
| 1554 |
+
{
|
| 1555 |
+
"epoch": 0.097,
|
| 1556 |
+
"eval_cos_sim": 0.8808521628379822,
|
| 1557 |
+
"eval_loss": 0.1201059584830947,
|
| 1558 |
+
"eval_runtime": 161.1408,
|
| 1559 |
+
"eval_samples_per_second": 24.823,
|
| 1560 |
+
"eval_steps_per_second": 0.391,
|
| 1561 |
+
"step": 970
|
| 1562 |
+
}
|
| 1563 |
+
],
|
| 1564 |
+
"logging_steps": 10,
|
| 1565 |
+
"max_steps": 10000,
|
| 1566 |
+
"num_input_tokens_seen": 0,
|
| 1567 |
+
"num_train_epochs": 9223372036854775807,
|
| 1568 |
+
"save_steps": 10,
|
| 1569 |
+
"stateful_callbacks": {
|
| 1570 |
+
"TrainerControl": {
|
| 1571 |
+
"args": {
|
| 1572 |
+
"should_epoch_stop": false,
|
| 1573 |
+
"should_evaluate": false,
|
| 1574 |
+
"should_log": false,
|
| 1575 |
+
"should_save": true,
|
| 1576 |
+
"should_training_stop": false
|
| 1577 |
+
},
|
| 1578 |
+
"attributes": {}
|
| 1579 |
+
}
|
| 1580 |
+
},
|
| 1581 |
+
"total_flos": 0.0,
|
| 1582 |
+
"train_batch_size": 440,
|
| 1583 |
+
"trial_name": null,
|
| 1584 |
+
"trial_params": null
|
| 1585 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c5ffeff2858a823619b1d15e1ef15f56c829886ecd33d46ec591b342d231b711
|
| 3 |
+
size 5176
|