Upload model
Browse files- best_results.json +13 -0
- model.safetensors +1 -1
- training_history.json +2496 -0
- training_state.pth +3 -0
best_results.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"sts_spearman": 0.7992766546129851,
|
| 3 |
+
"sts_pearson": 0.7723631373159269,
|
| 4 |
+
"retrieval_recall_at_1": 0.668,
|
| 5 |
+
"retrieval_recall_at_5": 0.936,
|
| 6 |
+
"retrieval_recall_at_10": 0.976,
|
| 7 |
+
"nli_accuracy": 0.5,
|
| 8 |
+
"nli_similarity": 0.7214555740356445,
|
| 9 |
+
"paraphrase_accuracy": 0.5,
|
| 10 |
+
"paraphrase_f1": 0.6666666666666666,
|
| 11 |
+
"paraphrase_similarity": 0.8633242845535278,
|
| 12 |
+
"composite_score": 0.7971049939731593
|
| 13 |
+
}
|
model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4082832
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d50a170ba9a88a50f8b72b92946ad32211f1d42f16bee9abb8a7019f3e44f8f8
|
| 3 |
size 4082832
|
training_history.json
ADDED
|
@@ -0,0 +1,2496 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"epoch": 1,
|
| 4 |
+
"eval_results": {
|
| 5 |
+
"sts_spearman": 0.8289338745472172,
|
| 6 |
+
"sts_pearson": 0.8098611426918698,
|
| 7 |
+
"retrieval_recall_at_1": 0.428,
|
| 8 |
+
"retrieval_recall_at_5": 0.746,
|
| 9 |
+
"retrieval_recall_at_10": 0.85,
|
| 10 |
+
"nli_accuracy": 0.5,
|
| 11 |
+
"nli_similarity": 0.7193673849105835,
|
| 12 |
+
"paraphrase_accuracy": 0.5,
|
| 13 |
+
"paraphrase_f1": 0.6666666666666666,
|
| 14 |
+
"paraphrase_similarity": 0.8657987713813782,
|
| 15 |
+
"composite_score": 0.7549336039402753
|
| 16 |
+
},
|
| 17 |
+
"losses": {
|
| 18 |
+
"sts": {
|
| 19 |
+
"total": 0.09194075802098149,
|
| 20 |
+
"distill": 0.07957400041429893,
|
| 21 |
+
"task": 0.2431019875018493,
|
| 22 |
+
"count": 23
|
| 23 |
+
},
|
| 24 |
+
"retrieval": {
|
| 25 |
+
"total": 0.268772747288359,
|
| 26 |
+
"distill": 0.0272021861945061,
|
| 27 |
+
"task": 1.241009996292439,
|
| 28 |
+
"count": 47
|
| 29 |
+
},
|
| 30 |
+
"nli": {
|
| 31 |
+
"total": 0.3314921881290192,
|
| 32 |
+
"distill": 0.01526121862549731,
|
| 33 |
+
"task": 2.3350987434387207,
|
| 34 |
+
"count": 47
|
| 35 |
+
},
|
| 36 |
+
"paraphrase": {
|
| 37 |
+
"total": 0.060653433203697205,
|
| 38 |
+
"distill": 0.008428953401744366,
|
| 39 |
+
"task": 0.8303535401821136,
|
| 40 |
+
"count": 10
|
| 41 |
+
}
|
| 42 |
+
},
|
| 43 |
+
"distill_weight": 0.3
|
| 44 |
+
},
|
| 45 |
+
{
|
| 46 |
+
"epoch": 2,
|
| 47 |
+
"eval_results": {
|
| 48 |
+
"sts_spearman": 0.8275112970530258,
|
| 49 |
+
"sts_pearson": 0.8086337285455611,
|
| 50 |
+
"retrieval_recall_at_1": 0.452,
|
| 51 |
+
"retrieval_recall_at_5": 0.78,
|
| 52 |
+
"retrieval_recall_at_10": 0.876,
|
| 53 |
+
"nli_accuracy": 0.5,
|
| 54 |
+
"nli_similarity": 0.7263807654380798,
|
| 55 |
+
"paraphrase_accuracy": 0.5,
|
| 56 |
+
"paraphrase_f1": 0.6666666666666666,
|
| 57 |
+
"paraphrase_similarity": 0.8608207106590271,
|
| 58 |
+
"composite_score": 0.7644223151931796
|
| 59 |
+
},
|
| 60 |
+
"losses": {
|
| 61 |
+
"sts": {
|
| 62 |
+
"total": 0.05613501726285271,
|
| 63 |
+
"distill": 0.00838314470551584,
|
| 64 |
+
"task": 0.19135420996209848,
|
| 65 |
+
"count": 23
|
| 66 |
+
},
|
| 67 |
+
"retrieval": {
|
| 68 |
+
"total": 0.21614952480539362,
|
| 69 |
+
"distill": 0.006375962016271784,
|
| 70 |
+
"task": 1.019319458210722,
|
| 71 |
+
"count": 47
|
| 72 |
+
},
|
| 73 |
+
"nli": {
|
| 74 |
+
"total": 0.30814098867964235,
|
| 75 |
+
"distill": 0.004959808186964786,
|
| 76 |
+
"task": 2.188524269043131,
|
| 77 |
+
"count": 47
|
| 78 |
+
},
|
| 79 |
+
"paraphrase": {
|
| 80 |
+
"total": 0.051446066424250605,
|
| 81 |
+
"distill": 0.0038678635377436877,
|
| 82 |
+
"task": 0.7177851617336273,
|
| 83 |
+
"count": 10
|
| 84 |
+
}
|
| 85 |
+
},
|
| 86 |
+
"distill_weight": 0.2994
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 3,
|
| 90 |
+
"eval_results": {
|
| 91 |
+
"sts_spearman": 0.827182891132439,
|
| 92 |
+
"sts_pearson": 0.809022237364101,
|
| 93 |
+
"retrieval_recall_at_1": 0.47,
|
| 94 |
+
"retrieval_recall_at_5": 0.8,
|
| 95 |
+
"retrieval_recall_at_10": 0.9,
|
| 96 |
+
"nli_accuracy": 0.5,
|
| 97 |
+
"nli_similarity": 0.7274820804595947,
|
| 98 |
+
"paraphrase_accuracy": 0.5,
|
| 99 |
+
"paraphrase_f1": 0.6666666666666666,
|
| 100 |
+
"paraphrase_similarity": 0.8593040704727173,
|
| 101 |
+
"composite_score": 0.7702581122328862
|
| 102 |
+
},
|
| 103 |
+
"losses": {
|
| 104 |
+
"sts": {
|
| 105 |
+
"total": 0.05474994104841481,
|
| 106 |
+
"distill": 0.0038318179709755855,
|
| 107 |
+
"task": 0.191118774206742,
|
| 108 |
+
"count": 23
|
| 109 |
+
},
|
| 110 |
+
"retrieval": {
|
| 111 |
+
"total": 0.1865766501807152,
|
| 112 |
+
"distill": 0.0033635680979870737,
|
| 113 |
+
"task": 0.8821620687525323,
|
| 114 |
+
"count": 47
|
| 115 |
+
},
|
| 116 |
+
"nli": {
|
| 117 |
+
"total": 0.2887607974574921,
|
| 118 |
+
"distill": 0.002587009018207801,
|
| 119 |
+
"task": 2.0535353625074344,
|
| 120 |
+
"count": 47
|
| 121 |
+
},
|
| 122 |
+
"paraphrase": {
|
| 123 |
+
"total": 0.04753798432648182,
|
| 124 |
+
"distill": 0.0023388142231851815,
|
| 125 |
+
"task": 0.6679855525493622,
|
| 126 |
+
"count": 10
|
| 127 |
+
}
|
| 128 |
+
},
|
| 129 |
+
"distill_weight": 0.2988
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 4,
|
| 133 |
+
"eval_results": {
|
| 134 |
+
"sts_spearman": 0.8273207309309314,
|
| 135 |
+
"sts_pearson": 0.8086748676612024,
|
| 136 |
+
"retrieval_recall_at_1": 0.496,
|
| 137 |
+
"retrieval_recall_at_5": 0.822,
|
| 138 |
+
"retrieval_recall_at_10": 0.914,
|
| 139 |
+
"nli_accuracy": 0.5,
|
| 140 |
+
"nli_similarity": 0.7310156226158142,
|
| 141 |
+
"paraphrase_accuracy": 0.5,
|
| 142 |
+
"paraphrase_f1": 0.6666666666666666,
|
| 143 |
+
"paraphrase_similarity": 0.8595395684242249,
|
| 144 |
+
"composite_score": 0.7769270321321324
|
| 145 |
+
},
|
| 146 |
+
"losses": {
|
| 147 |
+
"sts": {
|
| 148 |
+
"total": 0.05435633513590564,
|
| 149 |
+
"distill": 0.0022464350635266824,
|
| 150 |
+
"task": 0.19124554292015408,
|
| 151 |
+
"count": 23
|
| 152 |
+
},
|
| 153 |
+
"retrieval": {
|
| 154 |
+
"total": 0.16951107297171938,
|
| 155 |
+
"distill": 0.0021174795668017356,
|
| 156 |
+
"task": 0.8021261704728958,
|
| 157 |
+
"count": 47
|
| 158 |
+
},
|
| 159 |
+
"nli": {
|
| 160 |
+
"total": 0.280746913019647,
|
| 161 |
+
"distill": 0.0015788413361309374,
|
| 162 |
+
"task": 1.9968374790029322,
|
| 163 |
+
"count": 47
|
| 164 |
+
},
|
| 165 |
+
"paraphrase": {
|
| 166 |
+
"total": 0.044596540927886966,
|
| 167 |
+
"distill": 0.0015972736524417996,
|
| 168 |
+
"task": 0.628672468662262,
|
| 169 |
+
"count": 10
|
| 170 |
+
}
|
| 171 |
+
},
|
| 172 |
+
"distill_weight": 0.29819999999999997
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"epoch": 5,
|
| 176 |
+
"eval_results": {
|
| 177 |
+
"sts_spearman": 0.8252767207778248,
|
| 178 |
+
"sts_pearson": 0.8062386165350521,
|
| 179 |
+
"retrieval_recall_at_1": 0.51,
|
| 180 |
+
"retrieval_recall_at_5": 0.828,
|
| 181 |
+
"retrieval_recall_at_10": 0.922,
|
| 182 |
+
"nli_accuracy": 0.5,
|
| 183 |
+
"nli_similarity": 0.7306257486343384,
|
| 184 |
+
"paraphrase_accuracy": 0.5,
|
| 185 |
+
"paraphrase_f1": 0.6666666666666666,
|
| 186 |
+
"paraphrase_similarity": 0.8570234775543213,
|
| 187 |
+
"composite_score": 0.7777050270555791
|
| 188 |
+
},
|
| 189 |
+
"losses": {
|
| 190 |
+
"sts": {
|
| 191 |
+
"total": 0.05338770879999451,
|
| 192 |
+
"distill": 0.0014984115867106163,
|
| 193 |
+
"task": 0.1884317449901415,
|
| 194 |
+
"count": 23
|
| 195 |
+
},
|
| 196 |
+
"retrieval": {
|
| 197 |
+
"total": 0.15480095338314137,
|
| 198 |
+
"distill": 0.0014994340983437414,
|
| 199 |
+
"task": 0.7325110093076178,
|
| 200 |
+
"count": 47
|
| 201 |
+
},
|
| 202 |
+
"nli": {
|
| 203 |
+
"total": 0.27232939100011866,
|
| 204 |
+
"distill": 0.0010938003134140944,
|
| 205 |
+
"task": 1.936246319020048,
|
| 206 |
+
"count": 47
|
| 207 |
+
},
|
| 208 |
+
"paraphrase": {
|
| 209 |
+
"total": 0.04112741462886334,
|
| 210 |
+
"distill": 0.0012359182350337506,
|
| 211 |
+
"task": 0.5802905261516571,
|
| 212 |
+
"count": 10
|
| 213 |
+
}
|
| 214 |
+
},
|
| 215 |
+
"distill_weight": 0.2976
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 6,
|
| 219 |
+
"eval_results": {
|
| 220 |
+
"sts_spearman": 0.8248752970132536,
|
| 221 |
+
"sts_pearson": 0.805502215995378,
|
| 222 |
+
"retrieval_recall_at_1": 0.524,
|
| 223 |
+
"retrieval_recall_at_5": 0.83,
|
| 224 |
+
"retrieval_recall_at_10": 0.922,
|
| 225 |
+
"nli_accuracy": 0.5,
|
| 226 |
+
"nli_similarity": 0.731809675693512,
|
| 227 |
+
"paraphrase_accuracy": 0.5,
|
| 228 |
+
"paraphrase_f1": 0.6666666666666666,
|
| 229 |
+
"paraphrase_similarity": 0.8566122651100159,
|
| 230 |
+
"composite_score": 0.7781043151732935
|
| 231 |
+
},
|
| 232 |
+
"losses": {
|
| 233 |
+
"sts": {
|
| 234 |
+
"total": 0.05317179696715396,
|
| 235 |
+
"distill": 0.0011427759790145185,
|
| 236 |
+
"task": 0.1878819122262623,
|
| 237 |
+
"count": 23
|
| 238 |
+
},
|
| 239 |
+
"retrieval": {
|
| 240 |
+
"total": 0.14514635503292084,
|
| 241 |
+
"distill": 0.001184225406874209,
|
| 242 |
+
"task": 0.6865559283723223,
|
| 243 |
+
"count": 47
|
| 244 |
+
},
|
| 245 |
+
"nli": {
|
| 246 |
+
"total": 0.2656283996840741,
|
| 247 |
+
"distill": 0.0008388215606596242,
|
| 248 |
+
"task": 1.8874770808727184,
|
| 249 |
+
"count": 47
|
| 250 |
+
},
|
| 251 |
+
"paraphrase": {
|
| 252 |
+
"total": 0.038727284595370295,
|
| 253 |
+
"distill": 0.0010345598682761192,
|
| 254 |
+
"task": 0.546515229344368,
|
| 255 |
+
"count": 10
|
| 256 |
+
}
|
| 257 |
+
},
|
| 258 |
+
"distill_weight": 0.297
|
| 259 |
+
},
|
| 260 |
+
{
|
| 261 |
+
"epoch": 7,
|
| 262 |
+
"eval_results": {
|
| 263 |
+
"sts_spearman": 0.82402492098156,
|
| 264 |
+
"sts_pearson": 0.8039468293688954,
|
| 265 |
+
"retrieval_recall_at_1": 0.538,
|
| 266 |
+
"retrieval_recall_at_5": 0.842,
|
| 267 |
+
"retrieval_recall_at_10": 0.926,
|
| 268 |
+
"nli_accuracy": 0.5,
|
| 269 |
+
"nli_similarity": 0.7345654368400574,
|
| 270 |
+
"paraphrase_accuracy": 0.5,
|
| 271 |
+
"paraphrase_f1": 0.6666666666666666,
|
| 272 |
+
"paraphrase_similarity": 0.8579738140106201,
|
| 273 |
+
"composite_score": 0.7812791271574466
|
| 274 |
+
},
|
| 275 |
+
"losses": {
|
| 276 |
+
"sts": {
|
| 277 |
+
"total": 0.052247073501348495,
|
| 278 |
+
"distill": 0.0009337503781906613,
|
| 279 |
+
"task": 0.18465858309165292,
|
| 280 |
+
"count": 23
|
| 281 |
+
},
|
| 282 |
+
"retrieval": {
|
| 283 |
+
"total": 0.13885294297274123,
|
| 284 |
+
"distill": 0.0010064611387280548,
|
| 285 |
+
"task": 0.6564081304884971,
|
| 286 |
+
"count": 47
|
| 287 |
+
},
|
| 288 |
+
"nli": {
|
| 289 |
+
"total": 0.2596762865147692,
|
| 290 |
+
"distill": 0.0006988341003616756,
|
| 291 |
+
"task": 1.8438684280882491,
|
| 292 |
+
"count": 47
|
| 293 |
+
},
|
| 294 |
+
"paraphrase": {
|
| 295 |
+
"total": 0.03767538573592901,
|
| 296 |
+
"distill": 0.0009226226538885385,
|
| 297 |
+
"task": 0.5315793305635452,
|
| 298 |
+
"count": 10
|
| 299 |
+
}
|
| 300 |
+
},
|
| 301 |
+
"distill_weight": 0.2964
|
| 302 |
+
},
|
| 303 |
+
{
|
| 304 |
+
"epoch": 8,
|
| 305 |
+
"eval_results": {
|
| 306 |
+
"sts_spearman": 0.823563554115123,
|
| 307 |
+
"sts_pearson": 0.8030950418925904,
|
| 308 |
+
"retrieval_recall_at_1": 0.558,
|
| 309 |
+
"retrieval_recall_at_5": 0.85,
|
| 310 |
+
"retrieval_recall_at_10": 0.936,
|
| 311 |
+
"nli_accuracy": 0.5,
|
| 312 |
+
"nli_similarity": 0.7343195080757141,
|
| 313 |
+
"paraphrase_accuracy": 0.5,
|
| 314 |
+
"paraphrase_f1": 0.6666666666666666,
|
| 315 |
+
"paraphrase_similarity": 0.8582324385643005,
|
| 316 |
+
"composite_score": 0.7834484437242282
|
| 317 |
+
},
|
| 318 |
+
"losses": {
|
| 319 |
+
"sts": {
|
| 320 |
+
"total": 0.05065638410008472,
|
| 321 |
+
"distill": 0.0008207539668428185,
|
| 322 |
+
"task": 0.1789747496014056,
|
| 323 |
+
"count": 23
|
| 324 |
+
},
|
| 325 |
+
"retrieval": {
|
| 326 |
+
"total": 0.13189882991161753,
|
| 327 |
+
"distill": 0.0008978993072115044,
|
| 328 |
+
"task": 0.6230863763930949,
|
| 329 |
+
"count": 47
|
| 330 |
+
},
|
| 331 |
+
"nli": {
|
| 332 |
+
"total": 0.2552652054644646,
|
| 333 |
+
"distill": 0.0006161105479708218,
|
| 334 |
+
"task": 1.8111542945212507,
|
| 335 |
+
"count": 47
|
| 336 |
+
},
|
| 337 |
+
"paraphrase": {
|
| 338 |
+
"total": 0.03772428296506405,
|
| 339 |
+
"distill": 0.0008593377890065313,
|
| 340 |
+
"task": 0.5320944607257843,
|
| 341 |
+
"count": 10
|
| 342 |
+
}
|
| 343 |
+
},
|
| 344 |
+
"distill_weight": 0.2958
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 9,
|
| 348 |
+
"eval_results": {
|
| 349 |
+
"sts_spearman": 0.8224264234371386,
|
| 350 |
+
"sts_pearson": 0.801554516870732,
|
| 351 |
+
"retrieval_recall_at_1": 0.57,
|
| 352 |
+
"retrieval_recall_at_5": 0.86,
|
| 353 |
+
"retrieval_recall_at_10": 0.94,
|
| 354 |
+
"nli_accuracy": 0.5,
|
| 355 |
+
"nli_similarity": 0.733967125415802,
|
| 356 |
+
"paraphrase_accuracy": 0.5,
|
| 357 |
+
"paraphrase_f1": 0.6666666666666666,
|
| 358 |
+
"paraphrase_similarity": 0.8588850498199463,
|
| 359 |
+
"composite_score": 0.785879878385236
|
| 360 |
+
},
|
| 361 |
+
"losses": {
|
| 362 |
+
"sts": {
|
| 363 |
+
"total": 0.049955147601988006,
|
| 364 |
+
"distill": 0.0007517727663861992,
|
| 365 |
+
"task": 0.17640900611877441,
|
| 366 |
+
"count": 23
|
| 367 |
+
},
|
| 368 |
+
"retrieval": {
|
| 369 |
+
"total": 0.12419923918044314,
|
| 370 |
+
"distill": 0.0008363095508452426,
|
| 371 |
+
"task": 0.5862294866683635,
|
| 372 |
+
"count": 47
|
| 373 |
+
},
|
| 374 |
+
"nli": {
|
| 375 |
+
"total": 0.2503235790323704,
|
| 376 |
+
"distill": 0.0005653487615029704,
|
| 377 |
+
"task": 1.7746644527354138,
|
| 378 |
+
"count": 47
|
| 379 |
+
},
|
| 380 |
+
"paraphrase": {
|
| 381 |
+
"total": 0.035123253054916856,
|
| 382 |
+
"distill": 0.0008159744320437312,
|
| 383 |
+
"task": 0.49492592811584474,
|
| 384 |
+
"count": 10
|
| 385 |
+
}
|
| 386 |
+
},
|
| 387 |
+
"distill_weight": 0.2952
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 10,
|
| 391 |
+
"eval_results": {
|
| 392 |
+
"sts_spearman": 0.8213443657508973,
|
| 393 |
+
"sts_pearson": 0.8000685657254155,
|
| 394 |
+
"retrieval_recall_at_1": 0.574,
|
| 395 |
+
"retrieval_recall_at_5": 0.876,
|
| 396 |
+
"retrieval_recall_at_10": 0.942,
|
| 397 |
+
"nli_accuracy": 0.5,
|
| 398 |
+
"nli_similarity": 0.7336843609809875,
|
| 399 |
+
"paraphrase_accuracy": 0.5,
|
| 400 |
+
"paraphrase_f1": 0.6666666666666666,
|
| 401 |
+
"paraphrase_similarity": 0.8580488562583923,
|
| 402 |
+
"composite_score": 0.7901388495421153
|
| 403 |
+
},
|
| 404 |
+
"losses": {
|
| 405 |
+
"sts": {
|
| 406 |
+
"total": 0.050120412493529526,
|
| 407 |
+
"distill": 0.0007037564241529807,
|
| 408 |
+
"task": 0.1768963829330776,
|
| 409 |
+
"count": 23
|
| 410 |
+
},
|
| 411 |
+
"retrieval": {
|
| 412 |
+
"total": 0.11885773881952813,
|
| 413 |
+
"distill": 0.0007918142426283436,
|
| 414 |
+
"task": 0.5605541461325706,
|
| 415 |
+
"count": 47
|
| 416 |
+
},
|
| 417 |
+
"nli": {
|
| 418 |
+
"total": 0.24667784539943047,
|
| 419 |
+
"distill": 0.0005330673104370052,
|
| 420 |
+
"task": 1.7473829781755488,
|
| 421 |
+
"count": 47
|
| 422 |
+
},
|
| 423 |
+
"paraphrase": {
|
| 424 |
+
"total": 0.03359868098050356,
|
| 425 |
+
"distill": 0.0007893668138422072,
|
| 426 |
+
"task": 0.47301009893417356,
|
| 427 |
+
"count": 10
|
| 428 |
+
}
|
| 429 |
+
},
|
| 430 |
+
"distill_weight": 0.29460000000000003
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 11,
|
| 434 |
+
"eval_results": {
|
| 435 |
+
"sts_spearman": 0.8200398862029752,
|
| 436 |
+
"sts_pearson": 0.7984632150530062,
|
| 437 |
+
"retrieval_recall_at_1": 0.574,
|
| 438 |
+
"retrieval_recall_at_5": 0.88,
|
| 439 |
+
"retrieval_recall_at_10": 0.948,
|
| 440 |
+
"nli_accuracy": 0.5,
|
| 441 |
+
"nli_similarity": 0.7349022030830383,
|
| 442 |
+
"paraphrase_accuracy": 0.5,
|
| 443 |
+
"paraphrase_f1": 0.6666666666666666,
|
| 444 |
+
"paraphrase_similarity": 0.8587563037872314,
|
| 445 |
+
"composite_score": 0.7906866097681543
|
| 446 |
+
},
|
| 447 |
+
"losses": {
|
| 448 |
+
"sts": {
|
| 449 |
+
"total": 0.049540048222179,
|
| 450 |
+
"distill": 0.0006752971321870775,
|
| 451 |
+
"task": 0.17472205732179724,
|
| 452 |
+
"count": 23
|
| 453 |
+
},
|
| 454 |
+
"retrieval": {
|
| 455 |
+
"total": 0.1152741235304386,
|
| 456 |
+
"distill": 0.0007609698123873231,
|
| 457 |
+
"task": 0.543203037469945,
|
| 458 |
+
"count": 47
|
| 459 |
+
},
|
| 460 |
+
"nli": {
|
| 461 |
+
"total": 0.2454570157730833,
|
| 462 |
+
"distill": 0.000510894448218986,
|
| 463 |
+
"task": 1.7373002945108618,
|
| 464 |
+
"count": 47
|
| 465 |
+
},
|
| 466 |
+
"paraphrase": {
|
| 467 |
+
"total": 0.032351000048220155,
|
| 468 |
+
"distill": 0.0007691650826018304,
|
| 469 |
+
"task": 0.4550263941287994,
|
| 470 |
+
"count": 10
|
| 471 |
+
}
|
| 472 |
+
},
|
| 473 |
+
"distill_weight": 0.29400000000000004
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 12,
|
| 477 |
+
"eval_results": {
|
| 478 |
+
"sts_spearman": 0.8193103945963425,
|
| 479 |
+
"sts_pearson": 0.7974055277518916,
|
| 480 |
+
"retrieval_recall_at_1": 0.594,
|
| 481 |
+
"retrieval_recall_at_5": 0.884,
|
| 482 |
+
"retrieval_recall_at_10": 0.948,
|
| 483 |
+
"nli_accuracy": 0.5,
|
| 484 |
+
"nli_similarity": 0.7340207695960999,
|
| 485 |
+
"paraphrase_accuracy": 0.5,
|
| 486 |
+
"paraphrase_f1": 0.6666666666666666,
|
| 487 |
+
"paraphrase_similarity": 0.8586063981056213,
|
| 488 |
+
"composite_score": 0.7915218639648379
|
| 489 |
+
},
|
| 490 |
+
"losses": {
|
| 491 |
+
"sts": {
|
| 492 |
+
"total": 0.048518418131963066,
|
| 493 |
+
"distill": 0.0006525755262650225,
|
| 494 |
+
"task": 0.1709841172332349,
|
| 495 |
+
"count": 23
|
| 496 |
+
},
|
| 497 |
+
"retrieval": {
|
| 498 |
+
"total": 0.11415320951887901,
|
| 499 |
+
"distill": 0.0007357201898133659,
|
| 500 |
+
"task": 0.5374910330518763,
|
| 501 |
+
"count": 47
|
| 502 |
+
},
|
| 503 |
+
"nli": {
|
| 504 |
+
"total": 0.24104384951134947,
|
| 505 |
+
"distill": 0.0004927666555654179,
|
| 506 |
+
"task": 1.704636736119047,
|
| 507 |
+
"count": 47
|
| 508 |
+
},
|
| 509 |
+
"paraphrase": {
|
| 510 |
+
"total": 0.031325037218630315,
|
| 511 |
+
"distill": 0.0007606724102515727,
|
| 512 |
+
"task": 0.44016211330890653,
|
| 513 |
+
"count": 10
|
| 514 |
+
}
|
| 515 |
+
},
|
| 516 |
+
"distill_weight": 0.2934
|
| 517 |
+
},
|
| 518 |
+
{
|
| 519 |
+
"epoch": 13,
|
| 520 |
+
"eval_results": {
|
| 521 |
+
"sts_spearman": 0.8188842839819943,
|
| 522 |
+
"sts_pearson": 0.7964862034231038,
|
| 523 |
+
"retrieval_recall_at_1": 0.596,
|
| 524 |
+
"retrieval_recall_at_5": 0.896,
|
| 525 |
+
"retrieval_recall_at_10": 0.948,
|
| 526 |
+
"nli_accuracy": 0.5,
|
| 527 |
+
"nli_similarity": 0.7331050634384155,
|
| 528 |
+
"paraphrase_accuracy": 0.5,
|
| 529 |
+
"paraphrase_f1": 0.6666666666666666,
|
| 530 |
+
"paraphrase_similarity": 0.8598613142967224,
|
| 531 |
+
"composite_score": 0.7949088086576639
|
| 532 |
+
},
|
| 533 |
+
"losses": {
|
| 534 |
+
"sts": {
|
| 535 |
+
"total": 0.047971061390379204,
|
| 536 |
+
"distill": 0.000634684164137782,
|
| 537 |
+
"task": 0.1689240109661351,
|
| 538 |
+
"count": 23
|
| 539 |
+
},
|
| 540 |
+
"retrieval": {
|
| 541 |
+
"total": 0.109825795159695,
|
| 542 |
+
"distill": 0.000719507896103599,
|
| 543 |
+
"task": 0.5166625177606623,
|
| 544 |
+
"count": 47
|
| 545 |
+
},
|
| 546 |
+
"nli": {
|
| 547 |
+
"total": 0.23657965850322804,
|
| 548 |
+
"distill": 0.00048042560786385326,
|
| 549 |
+
"task": 1.6716556980254802,
|
| 550 |
+
"count": 47
|
| 551 |
+
},
|
| 552 |
+
"paraphrase": {
|
| 553 |
+
"total": 0.030759319849312305,
|
| 554 |
+
"distill": 0.0007484613277483731,
|
| 555 |
+
"task": 0.43184628784656526,
|
| 556 |
+
"count": 10
|
| 557 |
+
}
|
| 558 |
+
},
|
| 559 |
+
"distill_weight": 0.2928
|
| 560 |
+
},
|
| 561 |
+
{
|
| 562 |
+
"epoch": 14,
|
| 563 |
+
"eval_results": {
|
| 564 |
+
"sts_spearman": 0.8170877062297013,
|
| 565 |
+
"sts_pearson": 0.794295936858975,
|
| 566 |
+
"retrieval_recall_at_1": 0.596,
|
| 567 |
+
"retrieval_recall_at_5": 0.896,
|
| 568 |
+
"retrieval_recall_at_10": 0.948,
|
| 569 |
+
"nli_accuracy": 0.5,
|
| 570 |
+
"nli_similarity": 0.7348426580429077,
|
| 571 |
+
"paraphrase_accuracy": 0.5,
|
| 572 |
+
"paraphrase_f1": 0.6666666666666666,
|
| 573 |
+
"paraphrase_similarity": 0.8593454957008362,
|
| 574 |
+
"composite_score": 0.7940105197815173
|
| 575 |
+
},
|
| 576 |
+
"losses": {
|
| 577 |
+
"sts": {
|
| 578 |
+
"total": 0.04733733427913293,
|
| 579 |
+
"distill": 0.0006224965605803806,
|
| 580 |
+
"task": 0.16655636870342752,
|
| 581 |
+
"count": 23
|
| 582 |
+
},
|
| 583 |
+
"retrieval": {
|
| 584 |
+
"total": 0.10522193921373245,
|
| 585 |
+
"distill": 0.0007066282906886586,
|
| 586 |
+
"task": 0.49456279772393247,
|
| 587 |
+
"count": 47
|
| 588 |
+
},
|
| 589 |
+
"nli": {
|
| 590 |
+
"total": 0.23459308768840545,
|
| 591 |
+
"distill": 0.000471937776037908,
|
| 592 |
+
"task": 1.6562247961125476,
|
| 593 |
+
"count": 47
|
| 594 |
+
},
|
| 595 |
+
"paraphrase": {
|
| 596 |
+
"total": 0.02917077410966158,
|
| 597 |
+
"distill": 0.0007398281595669687,
|
| 598 |
+
"task": 0.40907877683639526,
|
| 599 |
+
"count": 10
|
| 600 |
+
}
|
| 601 |
+
},
|
| 602 |
+
"distill_weight": 0.2922
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 15,
|
| 606 |
+
"eval_results": {
|
| 607 |
+
"sts_spearman": 0.8157736839705673,
|
| 608 |
+
"sts_pearson": 0.7925540172374534,
|
| 609 |
+
"retrieval_recall_at_1": 0.606,
|
| 610 |
+
"retrieval_recall_at_5": 0.894,
|
| 611 |
+
"retrieval_recall_at_10": 0.952,
|
| 612 |
+
"nli_accuracy": 0.5,
|
| 613 |
+
"nli_similarity": 0.7331643104553223,
|
| 614 |
+
"paraphrase_accuracy": 0.5,
|
| 615 |
+
"paraphrase_f1": 0.6666666666666666,
|
| 616 |
+
"paraphrase_similarity": 0.8580483794212341,
|
| 617 |
+
"composite_score": 0.7927535086519504
|
| 618 |
+
},
|
| 619 |
+
"losses": {
|
| 620 |
+
"sts": {
|
| 621 |
+
"total": 0.04638928920030594,
|
| 622 |
+
"distill": 0.0006136388072501059,
|
| 623 |
+
"task": 0.16308000878147458,
|
| 624 |
+
"count": 23
|
| 625 |
+
},
|
| 626 |
+
"retrieval": {
|
| 627 |
+
"total": 0.10471683328456068,
|
| 628 |
+
"distill": 0.000696288561696147,
|
| 629 |
+
"task": 0.49178333104924954,
|
| 630 |
+
"count": 47
|
| 631 |
+
},
|
| 632 |
+
"nli": {
|
| 633 |
+
"total": 0.23013398304898688,
|
| 634 |
+
"distill": 0.00046382398540253174,
|
| 635 |
+
"task": 1.6233676443708704,
|
| 636 |
+
"count": 47
|
| 637 |
+
},
|
| 638 |
+
"paraphrase": {
|
| 639 |
+
"total": 0.028761257790029048,
|
| 640 |
+
"distill": 0.0007339846633840352,
|
| 641 |
+
"task": 0.4029817461967468,
|
| 642 |
+
"count": 10
|
| 643 |
+
}
|
| 644 |
+
},
|
| 645 |
+
"distill_weight": 0.2916
|
| 646 |
+
},
|
| 647 |
+
{
|
| 648 |
+
"epoch": 16,
|
| 649 |
+
"eval_results": {
|
| 650 |
+
"sts_spearman": 0.8157617853858435,
|
| 651 |
+
"sts_pearson": 0.7923932840541673,
|
| 652 |
+
"retrieval_recall_at_1": 0.612,
|
| 653 |
+
"retrieval_recall_at_5": 0.898,
|
| 654 |
+
"retrieval_recall_at_10": 0.954,
|
| 655 |
+
"nli_accuracy": 0.5,
|
| 656 |
+
"nli_similarity": 0.7322084903717041,
|
| 657 |
+
"paraphrase_accuracy": 0.5,
|
| 658 |
+
"paraphrase_f1": 0.6666666666666666,
|
| 659 |
+
"paraphrase_similarity": 0.8596447706222534,
|
| 660 |
+
"composite_score": 0.7939475593595884
|
| 661 |
+
},
|
| 662 |
+
"losses": {
|
| 663 |
+
"sts": {
|
| 664 |
+
"total": 0.04621984566683355,
|
| 665 |
+
"distill": 0.0006054732543618783,
|
| 666 |
+
"task": 0.1623542036699212,
|
| 667 |
+
"count": 23
|
| 668 |
+
},
|
| 669 |
+
"retrieval": {
|
| 670 |
+
"total": 0.10093737583845219,
|
| 671 |
+
"distill": 0.0006863788065024989,
|
| 672 |
+
"task": 0.473613737111396,
|
| 673 |
+
"count": 47
|
| 674 |
+
},
|
| 675 |
+
"nli": {
|
| 676 |
+
"total": 0.22668360205406837,
|
| 677 |
+
"distill": 0.00045654376067141904,
|
| 678 |
+
"task": 1.597678039936309,
|
| 679 |
+
"count": 47
|
| 680 |
+
},
|
| 681 |
+
"paraphrase": {
|
| 682 |
+
"total": 0.02780891638249159,
|
| 683 |
+
"distill": 0.0007290024077519774,
|
| 684 |
+
"task": 0.3892352104187012,
|
| 685 |
+
"count": 10
|
| 686 |
+
}
|
| 687 |
+
},
|
| 688 |
+
"distill_weight": 0.29100000000000004
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 17,
|
| 692 |
+
"eval_results": {
|
| 693 |
+
"sts_spearman": 0.8138228214142917,
|
| 694 |
+
"sts_pearson": 0.7899716079426576,
|
| 695 |
+
"retrieval_recall_at_1": 0.61,
|
| 696 |
+
"retrieval_recall_at_5": 0.9,
|
| 697 |
+
"retrieval_recall_at_10": 0.952,
|
| 698 |
+
"nli_accuracy": 0.5,
|
| 699 |
+
"nli_similarity": 0.7340738773345947,
|
| 700 |
+
"paraphrase_accuracy": 0.5,
|
| 701 |
+
"paraphrase_f1": 0.6666666666666666,
|
| 702 |
+
"paraphrase_similarity": 0.8595521450042725,
|
| 703 |
+
"composite_score": 0.7935780773738126
|
| 704 |
+
},
|
| 705 |
+
"losses": {
|
| 706 |
+
"sts": {
|
| 707 |
+
"total": 0.04601614598346793,
|
| 708 |
+
"distill": 0.0006001048604957759,
|
| 709 |
+
"task": 0.1615060425322989,
|
| 710 |
+
"count": 23
|
| 711 |
+
},
|
| 712 |
+
"retrieval": {
|
| 713 |
+
"total": 0.09758770988976702,
|
| 714 |
+
"distill": 0.0006810528612596557,
|
| 715 |
+
"task": 0.4574874652192948,
|
| 716 |
+
"count": 47
|
| 717 |
+
},
|
| 718 |
+
"nli": {
|
| 719 |
+
"total": 0.22612765399699516,
|
| 720 |
+
"distill": 0.00045324116202190203,
|
| 721 |
+
"task": 1.5924184956449143,
|
| 722 |
+
"count": 47
|
| 723 |
+
},
|
| 724 |
+
"paraphrase": {
|
| 725 |
+
"total": 0.02633261661976576,
|
| 726 |
+
"distill": 0.0007273759925737977,
|
| 727 |
+
"task": 0.3681142464280128,
|
| 728 |
+
"count": 10
|
| 729 |
+
}
|
| 730 |
+
},
|
| 731 |
+
"distill_weight": 0.2904
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 18,
|
| 735 |
+
"eval_results": {
|
| 736 |
+
"sts_spearman": 0.8137941541347218,
|
| 737 |
+
"sts_pearson": 0.7900114938692636,
|
| 738 |
+
"retrieval_recall_at_1": 0.618,
|
| 739 |
+
"retrieval_recall_at_5": 0.902,
|
| 740 |
+
"retrieval_recall_at_10": 0.954,
|
| 741 |
+
"nli_accuracy": 0.5,
|
| 742 |
+
"nli_similarity": 0.7320045232772827,
|
| 743 |
+
"paraphrase_accuracy": 0.5,
|
| 744 |
+
"paraphrase_f1": 0.6666666666666666,
|
| 745 |
+
"paraphrase_similarity": 0.8595419526100159,
|
| 746 |
+
"composite_score": 0.7941637437340276
|
| 747 |
+
},
|
| 748 |
+
"losses": {
|
| 749 |
+
"sts": {
|
| 750 |
+
"total": 0.04508868280960166,
|
| 751 |
+
"distill": 0.0005957435021865303,
|
| 752 |
+
"task": 0.15811052076194598,
|
| 753 |
+
"count": 23
|
| 754 |
+
},
|
| 755 |
+
"retrieval": {
|
| 756 |
+
"total": 0.09446902795040861,
|
| 757 |
+
"distill": 0.0006735043134540319,
|
| 758 |
+
"task": 0.4424755630340982,
|
| 759 |
+
"count": 47
|
| 760 |
+
},
|
| 761 |
+
"nli": {
|
| 762 |
+
"total": 0.2214733213186264,
|
| 763 |
+
"distill": 0.00044834198977580255,
|
| 764 |
+
"task": 1.5583173239484747,
|
| 765 |
+
"count": 47
|
| 766 |
+
},
|
| 767 |
+
"paraphrase": {
|
| 768 |
+
"total": 0.026047103106975555,
|
| 769 |
+
"distill": 0.0007218799262773245,
|
| 770 |
+
"task": 0.36381163746118544,
|
| 771 |
+
"count": 10
|
| 772 |
+
}
|
| 773 |
+
},
|
| 774 |
+
"distill_weight": 0.2898
|
| 775 |
+
},
|
| 776 |
+
{
|
| 777 |
+
"epoch": 19,
|
| 778 |
+
"eval_results": {
|
| 779 |
+
"sts_spearman": 0.8131324437136556,
|
| 780 |
+
"sts_pearson": 0.7888827667888834,
|
| 781 |
+
"retrieval_recall_at_1": 0.622,
|
| 782 |
+
"retrieval_recall_at_5": 0.906,
|
| 783 |
+
"retrieval_recall_at_10": 0.954,
|
| 784 |
+
"nli_accuracy": 0.5,
|
| 785 |
+
"nli_similarity": 0.7326275110244751,
|
| 786 |
+
"paraphrase_accuracy": 0.5,
|
| 787 |
+
"paraphrase_f1": 0.6666666666666666,
|
| 788 |
+
"paraphrase_similarity": 0.8593710064888,
|
| 789 |
+
"composite_score": 0.7950328885234945
|
| 790 |
+
},
|
| 791 |
+
"losses": {
|
| 792 |
+
"sts": {
|
| 793 |
+
"total": 0.044820788761843804,
|
| 794 |
+
"distill": 0.0005908587026288328,
|
| 795 |
+
"task": 0.1570410553527915,
|
| 796 |
+
"count": 23
|
| 797 |
+
},
|
| 798 |
+
"retrieval": {
|
| 799 |
+
"total": 0.0947175715514954,
|
| 800 |
+
"distill": 0.0006690982015843087,
|
| 801 |
+
"task": 0.44327550555797335,
|
| 802 |
+
"count": 47
|
| 803 |
+
},
|
| 804 |
+
"nli": {
|
| 805 |
+
"total": 0.22196958103078476,
|
| 806 |
+
"distill": 0.00044533125981886654,
|
| 807 |
+
"task": 1.5605007993414046,
|
| 808 |
+
"count": 47
|
| 809 |
+
},
|
| 810 |
+
"paraphrase": {
|
| 811 |
+
"total": 0.02539514433592558,
|
| 812 |
+
"distill": 0.000717665534466505,
|
| 813 |
+
"task": 0.35435559451580045,
|
| 814 |
+
"count": 10
|
| 815 |
+
}
|
| 816 |
+
},
|
| 817 |
+
"distill_weight": 0.2892
|
| 818 |
+
},
|
| 819 |
+
{
|
| 820 |
+
"epoch": 20,
|
| 821 |
+
"eval_results": {
|
| 822 |
+
"sts_spearman": 0.8125557315718881,
|
| 823 |
+
"sts_pearson": 0.7883415732948076,
|
| 824 |
+
"retrieval_recall_at_1": 0.622,
|
| 825 |
+
"retrieval_recall_at_5": 0.906,
|
| 826 |
+
"retrieval_recall_at_10": 0.956,
|
| 827 |
+
"nli_accuracy": 0.5,
|
| 828 |
+
"nli_similarity": 0.7300553321838379,
|
| 829 |
+
"paraphrase_accuracy": 0.5,
|
| 830 |
+
"paraphrase_f1": 0.6666666666666666,
|
| 831 |
+
"paraphrase_similarity": 0.8589670062065125,
|
| 832 |
+
"composite_score": 0.7947445324526108
|
| 833 |
+
},
|
| 834 |
+
"losses": {
|
| 835 |
+
"sts": {
|
| 836 |
+
"total": 0.04461472562473753,
|
| 837 |
+
"distill": 0.0005882442093697255,
|
| 838 |
+
"task": 0.15618835778340048,
|
| 839 |
+
"count": 23
|
| 840 |
+
},
|
| 841 |
+
"retrieval": {
|
| 842 |
+
"total": 0.09131274277225454,
|
| 843 |
+
"distill": 0.0006642562456111959,
|
| 844 |
+
"task": 0.4269564062991041,
|
| 845 |
+
"count": 47
|
| 846 |
+
},
|
| 847 |
+
"nli": {
|
| 848 |
+
"total": 0.21616570017439254,
|
| 849 |
+
"distill": 0.0004426293698030504,
|
| 850 |
+
"task": 1.518400065442349,
|
| 851 |
+
"count": 47
|
| 852 |
+
},
|
| 853 |
+
"paraphrase": {
|
| 854 |
+
"total": 0.024884730763733386,
|
| 855 |
+
"distill": 0.0007182472327258438,
|
| 856 |
+
"task": 0.3468856424093246,
|
| 857 |
+
"count": 10
|
| 858 |
+
}
|
| 859 |
+
},
|
| 860 |
+
"distill_weight": 0.2886
|
| 861 |
+
},
|
| 862 |
+
{
|
| 863 |
+
"epoch": 21,
|
| 864 |
+
"eval_results": {
|
| 865 |
+
"sts_spearman": 0.8124046923217468,
|
| 866 |
+
"sts_pearson": 0.7879715870379221,
|
| 867 |
+
"retrieval_recall_at_1": 0.626,
|
| 868 |
+
"retrieval_recall_at_5": 0.908,
|
| 869 |
+
"retrieval_recall_at_10": 0.954,
|
| 870 |
+
"nli_accuracy": 0.5,
|
| 871 |
+
"nli_similarity": 0.7299230694770813,
|
| 872 |
+
"paraphrase_accuracy": 0.5,
|
| 873 |
+
"paraphrase_f1": 0.6666666666666666,
|
| 874 |
+
"paraphrase_similarity": 0.8608139157295227,
|
| 875 |
+
"composite_score": 0.7952690128275401
|
| 876 |
+
},
|
| 877 |
+
"losses": {
|
| 878 |
+
"sts": {
|
| 879 |
+
"total": 0.044788526128167694,
|
| 880 |
+
"distill": 0.0005869252971656945,
|
| 881 |
+
"task": 0.15666956746059915,
|
| 882 |
+
"count": 23
|
| 883 |
+
},
|
| 884 |
+
"retrieval": {
|
| 885 |
+
"total": 0.09129251508002585,
|
| 886 |
+
"distill": 0.0006613465486728447,
|
| 887 |
+
"task": 0.42650772028781,
|
| 888 |
+
"count": 47
|
| 889 |
+
},
|
| 890 |
+
"nli": {
|
| 891 |
+
"total": 0.21487902199968378,
|
| 892 |
+
"distill": 0.0004392804477631332,
|
| 893 |
+
"task": 1.5080934889773105,
|
| 894 |
+
"count": 47
|
| 895 |
+
},
|
| 896 |
+
"paraphrase": {
|
| 897 |
+
"total": 0.02408289248123765,
|
| 898 |
+
"distill": 0.0007130690268240869,
|
| 899 |
+
"task": 0.3353585585951805,
|
| 900 |
+
"count": 10
|
| 901 |
+
}
|
| 902 |
+
},
|
| 903 |
+
"distill_weight": 0.28800000000000003
|
| 904 |
+
},
|
| 905 |
+
{
|
| 906 |
+
"epoch": 22,
|
| 907 |
+
"eval_results": {
|
| 908 |
+
"sts_spearman": 0.8108396163396415,
|
| 909 |
+
"sts_pearson": 0.7859428479306386,
|
| 910 |
+
"retrieval_recall_at_1": 0.626,
|
| 911 |
+
"retrieval_recall_at_5": 0.91,
|
| 912 |
+
"retrieval_recall_at_10": 0.956,
|
| 913 |
+
"nli_accuracy": 0.5,
|
| 914 |
+
"nli_similarity": 0.7304312586784363,
|
| 915 |
+
"paraphrase_accuracy": 0.5,
|
| 916 |
+
"paraphrase_f1": 0.6666666666666666,
|
| 917 |
+
"paraphrase_similarity": 0.8598558306694031,
|
| 918 |
+
"composite_score": 0.7950864748364874
|
| 919 |
+
},
|
| 920 |
+
"losses": {
|
| 921 |
+
"sts": {
|
| 922 |
+
"total": 0.04427984421667845,
|
| 923 |
+
"distill": 0.0005824335992498243,
|
| 924 |
+
"task": 0.15475882071515787,
|
| 925 |
+
"count": 23
|
| 926 |
+
},
|
| 927 |
+
"retrieval": {
|
| 928 |
+
"total": 0.08749476669633642,
|
| 929 |
+
"distill": 0.0006578407458406179,
|
| 930 |
+
"task": 0.40839041484163163,
|
| 931 |
+
"count": 47
|
| 932 |
+
},
|
| 933 |
+
"nli": {
|
| 934 |
+
"total": 0.21218161253218956,
|
| 935 |
+
"distill": 0.0004366559572171103,
|
| 936 |
+
"task": 1.487904317835544,
|
| 937 |
+
"count": 47
|
| 938 |
+
},
|
| 939 |
+
"paraphrase": {
|
| 940 |
+
"total": 0.02473886413499713,
|
| 941 |
+
"distill": 0.0007142506423406303,
|
| 942 |
+
"task": 0.34428276121616364,
|
| 943 |
+
"count": 10
|
| 944 |
+
}
|
| 945 |
+
},
|
| 946 |
+
"distill_weight": 0.2874
|
| 947 |
+
},
|
| 948 |
+
{
|
| 949 |
+
"epoch": 23,
|
| 950 |
+
"eval_results": {
|
| 951 |
+
"sts_spearman": 0.8101206106743032,
|
| 952 |
+
"sts_pearson": 0.7852421682051318,
|
| 953 |
+
"retrieval_recall_at_1": 0.626,
|
| 954 |
+
"retrieval_recall_at_5": 0.91,
|
| 955 |
+
"retrieval_recall_at_10": 0.956,
|
| 956 |
+
"nli_accuracy": 0.5,
|
| 957 |
+
"nli_similarity": 0.7298892140388489,
|
| 958 |
+
"paraphrase_accuracy": 0.5,
|
| 959 |
+
"paraphrase_f1": 0.6666666666666666,
|
| 960 |
+
"paraphrase_similarity": 0.8599591255187988,
|
| 961 |
+
"composite_score": 0.7947269720038184
|
| 962 |
+
},
|
| 963 |
+
"losses": {
|
| 964 |
+
"sts": {
|
| 965 |
+
"total": 0.043497207696023193,
|
| 966 |
+
"distill": 0.0005803209623200414,
|
| 967 |
+
"task": 0.15188857848229614,
|
| 968 |
+
"count": 23
|
| 969 |
+
},
|
| 970 |
+
"retrieval": {
|
| 971 |
+
"total": 0.08651064019253914,
|
| 972 |
+
"distill": 0.0006545566213119062,
|
| 973 |
+
"task": 0.40345349781056666,
|
| 974 |
+
"count": 47
|
| 975 |
+
},
|
| 976 |
+
"nli": {
|
| 977 |
+
"total": 0.21238021647676508,
|
| 978 |
+
"distill": 0.00043464487014794126,
|
| 979 |
+
"task": 1.4880508184432983,
|
| 980 |
+
"count": 47
|
| 981 |
+
},
|
| 982 |
+
"paraphrase": {
|
| 983 |
+
"total": 0.022220892272889613,
|
| 984 |
+
"distill": 0.0007110251928679645,
|
| 985 |
+
"task": 0.30870682895183565,
|
| 986 |
+
"count": 10
|
| 987 |
+
}
|
| 988 |
+
},
|
| 989 |
+
"distill_weight": 0.2868
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 24,
|
| 993 |
+
"eval_results": {
|
| 994 |
+
"sts_spearman": 0.8090374451655163,
|
| 995 |
+
"sts_pearson": 0.783875236477,
|
| 996 |
+
"retrieval_recall_at_1": 0.64,
|
| 997 |
+
"retrieval_recall_at_5": 0.91,
|
| 998 |
+
"retrieval_recall_at_10": 0.956,
|
| 999 |
+
"nli_accuracy": 0.5,
|
| 1000 |
+
"nli_similarity": 0.7307232618331909,
|
| 1001 |
+
"paraphrase_accuracy": 0.5,
|
| 1002 |
+
"paraphrase_f1": 0.6666666666666666,
|
| 1003 |
+
"paraphrase_similarity": 0.8604580163955688,
|
| 1004 |
+
"composite_score": 0.7941853892494248
|
| 1005 |
+
},
|
| 1006 |
+
"losses": {
|
| 1007 |
+
"sts": {
|
| 1008 |
+
"total": 0.04361056018134822,
|
| 1009 |
+
"distill": 0.0005793894398147645,
|
| 1010 |
+
"task": 0.15216006111839545,
|
| 1011 |
+
"count": 23
|
| 1012 |
+
},
|
| 1013 |
+
"retrieval": {
|
| 1014 |
+
"total": 0.08273381509996475,
|
| 1015 |
+
"distill": 0.0006529860733512868,
|
| 1016 |
+
"task": 0.3854811432513785,
|
| 1017 |
+
"count": 47
|
| 1018 |
+
},
|
| 1019 |
+
"nli": {
|
| 1020 |
+
"total": 0.2092740849611607,
|
| 1021 |
+
"distill": 0.00043218842759589724,
|
| 1022 |
+
"task": 1.4650490512239172,
|
| 1023 |
+
"count": 47
|
| 1024 |
+
},
|
| 1025 |
+
"paraphrase": {
|
| 1026 |
+
"total": 0.02230049455538392,
|
| 1027 |
+
"distill": 0.0007111194601748139,
|
| 1028 |
+
"task": 0.30956813097000124,
|
| 1029 |
+
"count": 10
|
| 1030 |
+
}
|
| 1031 |
+
},
|
| 1032 |
+
"distill_weight": 0.2862
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 25,
|
| 1036 |
+
"eval_results": {
|
| 1037 |
+
"sts_spearman": 0.8088290083594672,
|
| 1038 |
+
"sts_pearson": 0.7836111268009822,
|
| 1039 |
+
"retrieval_recall_at_1": 0.642,
|
| 1040 |
+
"retrieval_recall_at_5": 0.912,
|
| 1041 |
+
"retrieval_recall_at_10": 0.958,
|
| 1042 |
+
"nli_accuracy": 0.5,
|
| 1043 |
+
"nli_similarity": 0.7291015386581421,
|
| 1044 |
+
"paraphrase_accuracy": 0.5,
|
| 1045 |
+
"paraphrase_f1": 0.6666666666666666,
|
| 1046 |
+
"paraphrase_similarity": 0.8604010939598083,
|
| 1047 |
+
"composite_score": 0.7946811708464003
|
| 1048 |
+
},
|
| 1049 |
+
"losses": {
|
| 1050 |
+
"sts": {
|
| 1051 |
+
"total": 0.042825429821791855,
|
| 1052 |
+
"distill": 0.0005763423598735877,
|
| 1053 |
+
"task": 0.14928901389889096,
|
| 1054 |
+
"count": 23
|
| 1055 |
+
},
|
| 1056 |
+
"retrieval": {
|
| 1057 |
+
"total": 0.08428762876924048,
|
| 1058 |
+
"distill": 0.0006495943922113548,
|
| 1059 |
+
"task": 0.3924136973441915,
|
| 1060 |
+
"count": 47
|
| 1061 |
+
},
|
| 1062 |
+
"nli": {
|
| 1063 |
+
"total": 0.20725905958642352,
|
| 1064 |
+
"distill": 0.0004320865307508552,
|
| 1065 |
+
"task": 1.4497177093587024,
|
| 1066 |
+
"count": 47
|
| 1067 |
+
},
|
| 1068 |
+
"paraphrase": {
|
| 1069 |
+
"total": 0.021814299654215573,
|
| 1070 |
+
"distill": 0.0007105717668309808,
|
| 1071 |
+
"task": 0.30251066088676454,
|
| 1072 |
+
"count": 10
|
| 1073 |
+
}
|
| 1074 |
+
},
|
| 1075 |
+
"distill_weight": 0.2856
|
| 1076 |
+
},
|
| 1077 |
+
{
|
| 1078 |
+
"epoch": 26,
|
| 1079 |
+
"eval_results": {
|
| 1080 |
+
"sts_spearman": 0.8086444673068466,
|
| 1081 |
+
"sts_pearson": 0.7835424818834511,
|
| 1082 |
+
"retrieval_recall_at_1": 0.65,
|
| 1083 |
+
"retrieval_recall_at_5": 0.914,
|
| 1084 |
+
"retrieval_recall_at_10": 0.96,
|
| 1085 |
+
"nli_accuracy": 0.5,
|
| 1086 |
+
"nli_similarity": 0.7279387712478638,
|
| 1087 |
+
"paraphrase_accuracy": 0.5,
|
| 1088 |
+
"paraphrase_f1": 0.6666666666666666,
|
| 1089 |
+
"paraphrase_similarity": 0.8617146015167236,
|
| 1090 |
+
"composite_score": 0.79518890032009
|
| 1091 |
+
},
|
| 1092 |
+
"losses": {
|
| 1093 |
+
"sts": {
|
| 1094 |
+
"total": 0.04305816587546597,
|
| 1095 |
+
"distill": 0.0005753715294818192,
|
| 1096 |
+
"task": 0.14997966134029886,
|
| 1097 |
+
"count": 23
|
| 1098 |
+
},
|
| 1099 |
+
"retrieval": {
|
| 1100 |
+
"total": 0.08245786469667515,
|
| 1101 |
+
"distill": 0.0006471946550671883,
|
| 1102 |
+
"task": 0.3835590494439957,
|
| 1103 |
+
"count": 47
|
| 1104 |
+
},
|
| 1105 |
+
"nli": {
|
| 1106 |
+
"total": 0.20547382248208879,
|
| 1107 |
+
"distill": 0.00042902108806958225,
|
| 1108 |
+
"task": 1.4360247748963377,
|
| 1109 |
+
"count": 47
|
| 1110 |
+
},
|
| 1111 |
+
"paraphrase": {
|
| 1112 |
+
"total": 0.02213638899847865,
|
| 1113 |
+
"distill": 0.0007105463359039277,
|
| 1114 |
+
"task": 0.3067675843834877,
|
| 1115 |
+
"count": 10
|
| 1116 |
+
}
|
| 1117 |
+
},
|
| 1118 |
+
"distill_weight": 0.28500000000000003
|
| 1119 |
+
},
|
| 1120 |
+
{
|
| 1121 |
+
"epoch": 27,
|
| 1122 |
+
"eval_results": {
|
| 1123 |
+
"sts_spearman": 0.8075945769025906,
|
| 1124 |
+
"sts_pearson": 0.7817250896683322,
|
| 1125 |
+
"retrieval_recall_at_1": 0.648,
|
| 1126 |
+
"retrieval_recall_at_5": 0.916,
|
| 1127 |
+
"retrieval_recall_at_10": 0.964,
|
| 1128 |
+
"nli_accuracy": 0.5,
|
| 1129 |
+
"nli_similarity": 0.7290590405464172,
|
| 1130 |
+
"paraphrase_accuracy": 0.5,
|
| 1131 |
+
"paraphrase_f1": 0.6666666666666666,
|
| 1132 |
+
"paraphrase_similarity": 0.8608205318450928,
|
| 1133 |
+
"composite_score": 0.795263955117962
|
| 1134 |
+
},
|
| 1135 |
+
"losses": {
|
| 1136 |
+
"sts": {
|
| 1137 |
+
"total": 0.041979301558888474,
|
| 1138 |
+
"distill": 0.0005742409064070038,
|
| 1139 |
+
"task": 0.1460871498869813,
|
| 1140 |
+
"count": 23
|
| 1141 |
+
},
|
| 1142 |
+
"retrieval": {
|
| 1143 |
+
"total": 0.07854667179127957,
|
| 1144 |
+
"distill": 0.0006465072064918089,
|
| 1145 |
+
"task": 0.3650214504054252,
|
| 1146 |
+
"count": 47
|
| 1147 |
+
},
|
| 1148 |
+
"nli": {
|
| 1149 |
+
"total": 0.20393373991580718,
|
| 1150 |
+
"distill": 0.00042719431060485227,
|
| 1151 |
+
"task": 1.4240653768498848,
|
| 1152 |
+
"count": 47
|
| 1153 |
+
},
|
| 1154 |
+
"paraphrase": {
|
| 1155 |
+
"total": 0.02039573285728693,
|
| 1156 |
+
"distill": 0.0007079597911797464,
|
| 1157 |
+
"task": 0.28220218420028687,
|
| 1158 |
+
"count": 10
|
| 1159 |
+
}
|
| 1160 |
+
},
|
| 1161 |
+
"distill_weight": 0.2844
|
| 1162 |
+
},
|
| 1163 |
+
{
|
| 1164 |
+
"epoch": 28,
|
| 1165 |
+
"eval_results": {
|
| 1166 |
+
"sts_spearman": 0.8072111574266839,
|
| 1167 |
+
"sts_pearson": 0.7814816033631548,
|
| 1168 |
+
"retrieval_recall_at_1": 0.656,
|
| 1169 |
+
"retrieval_recall_at_5": 0.916,
|
| 1170 |
+
"retrieval_recall_at_10": 0.962,
|
| 1171 |
+
"nli_accuracy": 0.5,
|
| 1172 |
+
"nli_similarity": 0.7277703881263733,
|
| 1173 |
+
"paraphrase_accuracy": 0.5,
|
| 1174 |
+
"paraphrase_f1": 0.6666666666666666,
|
| 1175 |
+
"paraphrase_similarity": 0.8611229062080383,
|
| 1176 |
+
"composite_score": 0.7950722453800086
|
| 1177 |
+
},
|
| 1178 |
+
"losses": {
|
| 1179 |
+
"sts": {
|
| 1180 |
+
"total": 0.04250466094716736,
|
| 1181 |
+
"distill": 0.0005732332761196986,
|
| 1182 |
+
"task": 0.14780080836752188,
|
| 1183 |
+
"count": 23
|
| 1184 |
+
},
|
| 1185 |
+
"retrieval": {
|
| 1186 |
+
"total": 0.07828945341579457,
|
| 1187 |
+
"distill": 0.0006427260170234962,
|
| 1188 |
+
"task": 0.363525298681665,
|
| 1189 |
+
"count": 47
|
| 1190 |
+
},
|
| 1191 |
+
"nli": {
|
| 1192 |
+
"total": 0.20262091527593898,
|
| 1193 |
+
"distill": 0.0004252295209184051,
|
| 1194 |
+
"task": 1.4137128566173798,
|
| 1195 |
+
"count": 47
|
| 1196 |
+
},
|
| 1197 |
+
"paraphrase": {
|
| 1198 |
+
"total": 0.021627166587859393,
|
| 1199 |
+
"distill": 0.000706591084599495,
|
| 1200 |
+
"task": 0.29917111396789553,
|
| 1201 |
+
"count": 10
|
| 1202 |
+
}
|
| 1203 |
+
},
|
| 1204 |
+
"distill_weight": 0.2838
|
| 1205 |
+
},
|
| 1206 |
+
{
|
| 1207 |
+
"epoch": 29,
|
| 1208 |
+
"eval_results": {
|
| 1209 |
+
"sts_spearman": 0.8061150812196531,
|
| 1210 |
+
"sts_pearson": 0.7799263057126831,
|
| 1211 |
+
"retrieval_recall_at_1": 0.654,
|
| 1212 |
+
"retrieval_recall_at_5": 0.918,
|
| 1213 |
+
"retrieval_recall_at_10": 0.964,
|
| 1214 |
+
"nli_accuracy": 0.5,
|
| 1215 |
+
"nli_similarity": 0.7274843454360962,
|
| 1216 |
+
"paraphrase_accuracy": 0.5,
|
| 1217 |
+
"paraphrase_f1": 0.6666666666666666,
|
| 1218 |
+
"paraphrase_similarity": 0.8605352640151978,
|
| 1219 |
+
"composite_score": 0.7951242072764932
|
| 1220 |
+
},
|
| 1221 |
+
"losses": {
|
| 1222 |
+
"sts": {
|
| 1223 |
+
"total": 0.04178553026007569,
|
| 1224 |
+
"distill": 0.0005705209104749172,
|
| 1225 |
+
"task": 0.14517284346663434,
|
| 1226 |
+
"count": 23
|
| 1227 |
+
},
|
| 1228 |
+
"retrieval": {
|
| 1229 |
+
"total": 0.07806510272178244,
|
| 1230 |
+
"distill": 0.0006415697936701806,
|
| 1231 |
+
"task": 0.3621810408348733,
|
| 1232 |
+
"count": 47
|
| 1233 |
+
},
|
| 1234 |
+
"nli": {
|
| 1235 |
+
"total": 0.20029219636257659,
|
| 1236 |
+
"distill": 0.00042426467754263825,
|
| 1237 |
+
"task": 1.3962893257749842,
|
| 1238 |
+
"count": 47
|
| 1239 |
+
},
|
| 1240 |
+
"paraphrase": {
|
| 1241 |
+
"total": 0.020195775851607322,
|
| 1242 |
+
"distill": 0.000706263561733067,
|
| 1243 |
+
"task": 0.2789587274193764,
|
| 1244 |
+
"count": 10
|
| 1245 |
+
}
|
| 1246 |
+
},
|
| 1247 |
+
"distill_weight": 0.2832
|
| 1248 |
+
},
|
| 1249 |
+
{
|
| 1250 |
+
"epoch": 30,
|
| 1251 |
+
"eval_results": {
|
| 1252 |
+
"sts_spearman": 0.805952102459493,
|
| 1253 |
+
"sts_pearson": 0.7796381825168456,
|
| 1254 |
+
"retrieval_recall_at_1": 0.658,
|
| 1255 |
+
"retrieval_recall_at_5": 0.92,
|
| 1256 |
+
"retrieval_recall_at_10": 0.966,
|
| 1257 |
+
"nli_accuracy": 0.5,
|
| 1258 |
+
"nli_similarity": 0.7271153926849365,
|
| 1259 |
+
"paraphrase_accuracy": 0.5,
|
| 1260 |
+
"paraphrase_f1": 0.6666666666666666,
|
| 1261 |
+
"paraphrase_similarity": 0.8611077666282654,
|
| 1262 |
+
"composite_score": 0.7956427178964132
|
| 1263 |
+
},
|
| 1264 |
+
"losses": {
|
| 1265 |
+
"sts": {
|
| 1266 |
+
"total": 0.041891162486180016,
|
| 1267 |
+
"distill": 0.0005706739384154587,
|
| 1268 |
+
"task": 0.14542057948267978,
|
| 1269 |
+
"count": 23
|
| 1270 |
+
},
|
| 1271 |
+
"retrieval": {
|
| 1272 |
+
"total": 0.07668351699063118,
|
| 1273 |
+
"distill": 0.0006396458459463208,
|
| 1274 |
+
"task": 0.35546302129613594,
|
| 1275 |
+
"count": 47
|
| 1276 |
+
},
|
| 1277 |
+
"nli": {
|
| 1278 |
+
"total": 0.19826114177703857,
|
| 1279 |
+
"distill": 0.00042267177598253686,
|
| 1280 |
+
"task": 1.3809708052493157,
|
| 1281 |
+
"count": 47
|
| 1282 |
+
},
|
| 1283 |
+
"paraphrase": {
|
| 1284 |
+
"total": 0.019489498622715474,
|
| 1285 |
+
"distill": 0.0007044955214951188,
|
| 1286 |
+
"task": 0.268893338739872,
|
| 1287 |
+
"count": 10
|
| 1288 |
+
}
|
| 1289 |
+
},
|
| 1290 |
+
"distill_weight": 0.2826
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 31,
|
| 1294 |
+
"eval_results": {
|
| 1295 |
+
"sts_spearman": 0.8054395024616142,
|
| 1296 |
+
"sts_pearson": 0.7790870358568797,
|
| 1297 |
+
"retrieval_recall_at_1": 0.662,
|
| 1298 |
+
"retrieval_recall_at_5": 0.92,
|
| 1299 |
+
"retrieval_recall_at_10": 0.968,
|
| 1300 |
+
"nli_accuracy": 0.5,
|
| 1301 |
+
"nli_similarity": 0.7270572781562805,
|
| 1302 |
+
"paraphrase_accuracy": 0.5,
|
| 1303 |
+
"paraphrase_f1": 0.6666666666666666,
|
| 1304 |
+
"paraphrase_similarity": 0.8614012002944946,
|
| 1305 |
+
"composite_score": 0.7953864178974739
|
| 1306 |
+
},
|
| 1307 |
+
"losses": {
|
| 1308 |
+
"sts": {
|
| 1309 |
+
"total": 0.04180093708893527,
|
| 1310 |
+
"distill": 0.000569236811513648,
|
| 1311 |
+
"task": 0.14498750833065613,
|
| 1312 |
+
"count": 23
|
| 1313 |
+
},
|
| 1314 |
+
"retrieval": {
|
| 1315 |
+
"total": 0.07472586061092133,
|
| 1316 |
+
"distill": 0.0006379124688658308,
|
| 1317 |
+
"task": 0.346081572644254,
|
| 1318 |
+
"count": 47
|
| 1319 |
+
},
|
| 1320 |
+
"nli": {
|
| 1321 |
+
"total": 0.19744586469011105,
|
| 1322 |
+
"distill": 0.0004216810487745766,
|
| 1323 |
+
"task": 1.3741430931902947,
|
| 1324 |
+
"count": 47
|
| 1325 |
+
},
|
| 1326 |
+
"paraphrase": {
|
| 1327 |
+
"total": 0.02020758679136634,
|
| 1328 |
+
"distill": 0.0007039231946691871,
|
| 1329 |
+
"task": 0.27867799401283266,
|
| 1330 |
+
"count": 10
|
| 1331 |
+
}
|
| 1332 |
+
},
|
| 1333 |
+
"distill_weight": 0.28200000000000003
|
| 1334 |
+
},
|
| 1335 |
+
{
|
| 1336 |
+
"epoch": 32,
|
| 1337 |
+
"eval_results": {
|
| 1338 |
+
"sts_spearman": 0.8039734694214193,
|
| 1339 |
+
"sts_pearson": 0.7774542134736094,
|
| 1340 |
+
"retrieval_recall_at_1": 0.658,
|
| 1341 |
+
"retrieval_recall_at_5": 0.922,
|
| 1342 |
+
"retrieval_recall_at_10": 0.972,
|
| 1343 |
+
"nli_accuracy": 0.5,
|
| 1344 |
+
"nli_similarity": 0.7258086204528809,
|
| 1345 |
+
"paraphrase_accuracy": 0.5,
|
| 1346 |
+
"paraphrase_f1": 0.6666666666666666,
|
| 1347 |
+
"paraphrase_similarity": 0.8613040447235107,
|
| 1348 |
+
"composite_score": 0.7952534013773764
|
| 1349 |
+
},
|
| 1350 |
+
"losses": {
|
| 1351 |
+
"sts": {
|
| 1352 |
+
"total": 0.04136973813824032,
|
| 1353 |
+
"distill": 0.000568263144131102,
|
| 1354 |
+
"task": 0.14336845570284387,
|
| 1355 |
+
"count": 23
|
| 1356 |
+
},
|
| 1357 |
+
"retrieval": {
|
| 1358 |
+
"total": 0.07349338850125353,
|
| 1359 |
+
"distill": 0.000635797770921775,
|
| 1360 |
+
"task": 0.3400801315586618,
|
| 1361 |
+
"count": 47
|
| 1362 |
+
},
|
| 1363 |
+
"nli": {
|
| 1364 |
+
"total": 0.195538673629152,
|
| 1365 |
+
"distill": 0.00042041142513242333,
|
| 1366 |
+
"task": 1.3597298099639568,
|
| 1367 |
+
"count": 47
|
| 1368 |
+
},
|
| 1369 |
+
"paraphrase": {
|
| 1370 |
+
"total": 0.01802230104804039,
|
| 1371 |
+
"distill": 0.0007017374911811203,
|
| 1372 |
+
"task": 0.24804942756891252,
|
| 1373 |
+
"count": 10
|
| 1374 |
+
}
|
| 1375 |
+
},
|
| 1376 |
+
"distill_weight": 0.2814
|
| 1377 |
+
},
|
| 1378 |
+
{
|
| 1379 |
+
"epoch": 33,
|
| 1380 |
+
"eval_results": {
|
| 1381 |
+
"sts_spearman": 0.8034526879232323,
|
| 1382 |
+
"sts_pearson": 0.7767872504658838,
|
| 1383 |
+
"retrieval_recall_at_1": 0.664,
|
| 1384 |
+
"retrieval_recall_at_5": 0.92,
|
| 1385 |
+
"retrieval_recall_at_10": 0.972,
|
| 1386 |
+
"nli_accuracy": 0.5,
|
| 1387 |
+
"nli_similarity": 0.7267512083053589,
|
| 1388 |
+
"paraphrase_accuracy": 0.5,
|
| 1389 |
+
"paraphrase_f1": 0.6666666666666666,
|
| 1390 |
+
"paraphrase_similarity": 0.8609178066253662,
|
| 1391 |
+
"composite_score": 0.7943930106282828
|
| 1392 |
+
},
|
| 1393 |
+
"losses": {
|
| 1394 |
+
"sts": {
|
| 1395 |
+
"total": 0.04124606368334397,
|
| 1396 |
+
"distill": 0.0005672681145370007,
|
| 1397 |
+
"task": 0.14282110203867374,
|
| 1398 |
+
"count": 23
|
| 1399 |
+
},
|
| 1400 |
+
"retrieval": {
|
| 1401 |
+
"total": 0.07269588295132556,
|
| 1402 |
+
"distill": 0.0006344794367558937,
|
| 1403 |
+
"task": 0.3361036536541391,
|
| 1404 |
+
"count": 47
|
| 1405 |
+
},
|
| 1406 |
+
"nli": {
|
| 1407 |
+
"total": 0.19442298532800473,
|
| 1408 |
+
"distill": 0.0004187757124569505,
|
| 1409 |
+
"task": 1.3508439393753702,
|
| 1410 |
+
"count": 47
|
| 1411 |
+
},
|
| 1412 |
+
"paraphrase": {
|
| 1413 |
+
"total": 0.01792764011770487,
|
| 1414 |
+
"distill": 0.0007010513567365706,
|
| 1415 |
+
"task": 0.246534825861454,
|
| 1416 |
+
"count": 10
|
| 1417 |
+
}
|
| 1418 |
+
},
|
| 1419 |
+
"distill_weight": 0.2808
|
| 1420 |
+
},
|
| 1421 |
+
{
|
| 1422 |
+
"epoch": 34,
|
| 1423 |
+
"eval_results": {
|
| 1424 |
+
"sts_spearman": 0.8029147605057396,
|
| 1425 |
+
"sts_pearson": 0.7758699525760386,
|
| 1426 |
+
"retrieval_recall_at_1": 0.656,
|
| 1427 |
+
"retrieval_recall_at_5": 0.924,
|
| 1428 |
+
"retrieval_recall_at_10": 0.976,
|
| 1429 |
+
"nli_accuracy": 0.5,
|
| 1430 |
+
"nli_similarity": 0.7263163924217224,
|
| 1431 |
+
"paraphrase_accuracy": 0.5,
|
| 1432 |
+
"paraphrase_f1": 0.6666666666666666,
|
| 1433 |
+
"paraphrase_similarity": 0.861183762550354,
|
| 1434 |
+
"composite_score": 0.7953240469195365
|
| 1435 |
+
},
|
| 1436 |
+
"losses": {
|
| 1437 |
+
"sts": {
|
| 1438 |
+
"total": 0.04087469354271889,
|
| 1439 |
+
"distill": 0.0005659387705078268,
|
| 1440 |
+
"task": 0.14141468995291254,
|
| 1441 |
+
"count": 23
|
| 1442 |
+
},
|
| 1443 |
+
"retrieval": {
|
| 1444 |
+
"total": 0.07193009151106185,
|
| 1445 |
+
"distill": 0.0006315128511174562,
|
| 1446 |
+
"task": 0.3322827714554807,
|
| 1447 |
+
"count": 47
|
| 1448 |
+
},
|
| 1449 |
+
"nli": {
|
| 1450 |
+
"total": 0.19458080923303644,
|
| 1451 |
+
"distill": 0.0004175955335550169,
|
| 1452 |
+
"task": 1.3508182814780703,
|
| 1453 |
+
"count": 47
|
| 1454 |
+
},
|
| 1455 |
+
"paraphrase": {
|
| 1456 |
+
"total": 0.018750363681465387,
|
| 1457 |
+
"distill": 0.0007004878658335656,
|
| 1458 |
+
"task": 0.2577672630548477,
|
| 1459 |
+
"count": 10
|
| 1460 |
+
}
|
| 1461 |
+
},
|
| 1462 |
+
"distill_weight": 0.2802
|
| 1463 |
+
},
|
| 1464 |
+
{
|
| 1465 |
+
"epoch": 35,
|
| 1466 |
+
"eval_results": {
|
| 1467 |
+
"sts_spearman": 0.8029120487600959,
|
| 1468 |
+
"sts_pearson": 0.7759091259908452,
|
| 1469 |
+
"retrieval_recall_at_1": 0.652,
|
| 1470 |
+
"retrieval_recall_at_5": 0.924,
|
| 1471 |
+
"retrieval_recall_at_10": 0.976,
|
| 1472 |
+
"nli_accuracy": 0.5,
|
| 1473 |
+
"nli_similarity": 0.7263759970664978,
|
| 1474 |
+
"paraphrase_accuracy": 0.5,
|
| 1475 |
+
"paraphrase_f1": 0.6666666666666666,
|
| 1476 |
+
"paraphrase_similarity": 0.8618313670158386,
|
| 1477 |
+
"composite_score": 0.7953226910467146
|
| 1478 |
+
},
|
| 1479 |
+
"losses": {
|
| 1480 |
+
"sts": {
|
| 1481 |
+
"total": 0.040802220611468605,
|
| 1482 |
+
"distill": 0.0005649596442589941,
|
| 1483 |
+
"task": 0.1410475363549979,
|
| 1484 |
+
"count": 23
|
| 1485 |
+
},
|
| 1486 |
+
"retrieval": {
|
| 1487 |
+
"total": 0.07252203149998442,
|
| 1488 |
+
"distill": 0.0006298948205670619,
|
| 1489 |
+
"task": 0.33474880393515244,
|
| 1490 |
+
"count": 47
|
| 1491 |
+
},
|
| 1492 |
+
"nli": {
|
| 1493 |
+
"total": 0.19197505680804558,
|
| 1494 |
+
"distill": 0.0004152855129932311,
|
| 1495 |
+
"task": 1.3316139946592616,
|
| 1496 |
+
"count": 47
|
| 1497 |
+
},
|
| 1498 |
+
"paraphrase": {
|
| 1499 |
+
"total": 0.01693342700600624,
|
| 1500 |
+
"distill": 0.0006995417177677154,
|
| 1501 |
+
"task": 0.23234085589647294,
|
| 1502 |
+
"count": 10
|
| 1503 |
+
}
|
| 1504 |
+
},
|
| 1505 |
+
"distill_weight": 0.2796
|
| 1506 |
+
},
|
| 1507 |
+
{
|
| 1508 |
+
"epoch": 36,
|
| 1509 |
+
"eval_results": {
|
| 1510 |
+
"sts_spearman": 0.802407902504706,
|
| 1511 |
+
"sts_pearson": 0.7756569556734015,
|
| 1512 |
+
"retrieval_recall_at_1": 0.658,
|
| 1513 |
+
"retrieval_recall_at_5": 0.924,
|
| 1514 |
+
"retrieval_recall_at_10": 0.974,
|
| 1515 |
+
"nli_accuracy": 0.5,
|
| 1516 |
+
"nli_similarity": 0.7243511080741882,
|
| 1517 |
+
"paraphrase_accuracy": 0.5,
|
| 1518 |
+
"paraphrase_f1": 0.6666666666666666,
|
| 1519 |
+
"paraphrase_similarity": 0.8609747290611267,
|
| 1520 |
+
"composite_score": 0.7950706179190197
|
| 1521 |
+
},
|
| 1522 |
+
"losses": {
|
| 1523 |
+
"sts": {
|
| 1524 |
+
"total": 0.040637732200000595,
|
| 1525 |
+
"distill": 0.0005651830908154016,
|
| 1526 |
+
"task": 0.14036077325758728,
|
| 1527 |
+
"count": 23
|
| 1528 |
+
},
|
| 1529 |
+
"retrieval": {
|
| 1530 |
+
"total": 0.07090983135586089,
|
| 1531 |
+
"distill": 0.0006288531424596588,
|
| 1532 |
+
"task": 0.32701979894587335,
|
| 1533 |
+
"count": 47
|
| 1534 |
+
},
|
| 1535 |
+
"nli": {
|
| 1536 |
+
"total": 0.1879324754501911,
|
| 1537 |
+
"distill": 0.0004152300212770066,
|
| 1538 |
+
"task": 1.3024731620829155,
|
| 1539 |
+
"count": 47
|
| 1540 |
+
},
|
| 1541 |
+
"paraphrase": {
|
| 1542 |
+
"total": 0.01749492483213544,
|
| 1543 |
+
"distill": 0.0006997698335908353,
|
| 1544 |
+
"task": 0.23994021117687225,
|
| 1545 |
+
"count": 10
|
| 1546 |
+
}
|
| 1547 |
+
},
|
| 1548 |
+
"distill_weight": 0.279
|
| 1549 |
+
},
|
| 1550 |
+
{
|
| 1551 |
+
"epoch": 37,
|
| 1552 |
+
"eval_results": {
|
| 1553 |
+
"sts_spearman": 0.8014589498925283,
|
| 1554 |
+
"sts_pearson": 0.7745302218047226,
|
| 1555 |
+
"retrieval_recall_at_1": 0.662,
|
| 1556 |
+
"retrieval_recall_at_5": 0.926,
|
| 1557 |
+
"retrieval_recall_at_10": 0.974,
|
| 1558 |
+
"nli_accuracy": 0.5,
|
| 1559 |
+
"nli_similarity": 0.7247843742370605,
|
| 1560 |
+
"paraphrase_accuracy": 0.5,
|
| 1561 |
+
"paraphrase_f1": 0.6666666666666666,
|
| 1562 |
+
"paraphrase_similarity": 0.8621292114257812,
|
| 1563 |
+
"composite_score": 0.7951961416129308
|
| 1564 |
+
},
|
| 1565 |
+
"losses": {
|
| 1566 |
+
"sts": {
|
| 1567 |
+
"total": 0.0404453170688256,
|
| 1568 |
+
"distill": 0.0005647512937329062,
|
| 1569 |
+
"task": 0.13957903113054193,
|
| 1570 |
+
"count": 23
|
| 1571 |
+
},
|
| 1572 |
+
"retrieval": {
|
| 1573 |
+
"total": 0.07041878134329269,
|
| 1574 |
+
"distill": 0.000626888732980699,
|
| 1575 |
+
"task": 0.3244838112212242,
|
| 1576 |
+
"count": 47
|
| 1577 |
+
},
|
| 1578 |
+
"nli": {
|
| 1579 |
+
"total": 0.18824834677767247,
|
| 1580 |
+
"distill": 0.00041336335856071175,
|
| 1581 |
+
"task": 1.3035842015388164,
|
| 1582 |
+
"count": 47
|
| 1583 |
+
},
|
| 1584 |
+
"paraphrase": {
|
| 1585 |
+
"total": 0.017420530039817094,
|
| 1586 |
+
"distill": 0.0006980633072089404,
|
| 1587 |
+
"task": 0.23872213810682297,
|
| 1588 |
+
"count": 10
|
| 1589 |
+
}
|
| 1590 |
+
},
|
| 1591 |
+
"distill_weight": 0.2784
|
| 1592 |
+
},
|
| 1593 |
+
{
|
| 1594 |
+
"epoch": 38,
|
| 1595 |
+
"eval_results": {
|
| 1596 |
+
"sts_spearman": 0.8015481972850798,
|
| 1597 |
+
"sts_pearson": 0.7748134816243155,
|
| 1598 |
+
"retrieval_recall_at_1": 0.662,
|
| 1599 |
+
"retrieval_recall_at_5": 0.926,
|
| 1600 |
+
"retrieval_recall_at_10": 0.974,
|
| 1601 |
+
"nli_accuracy": 0.5,
|
| 1602 |
+
"nli_similarity": 0.7238280177116394,
|
| 1603 |
+
"paraphrase_accuracy": 0.5,
|
| 1604 |
+
"paraphrase_f1": 0.6666666666666666,
|
| 1605 |
+
"paraphrase_similarity": 0.8617278337478638,
|
| 1606 |
+
"composite_score": 0.7952407653092066
|
| 1607 |
+
},
|
| 1608 |
+
"losses": {
|
| 1609 |
+
"sts": {
|
| 1610 |
+
"total": 0.040147524246055145,
|
| 1611 |
+
"distill": 0.000563749264034888,
|
| 1612 |
+
"task": 0.13843435340601465,
|
| 1613 |
+
"count": 23
|
| 1614 |
+
},
|
| 1615 |
+
"retrieval": {
|
| 1616 |
+
"total": 0.06983992354349887,
|
| 1617 |
+
"distill": 0.0006249398723779682,
|
| 1618 |
+
"task": 0.3215467447930194,
|
| 1619 |
+
"count": 47
|
| 1620 |
+
},
|
| 1621 |
+
"nli": {
|
| 1622 |
+
"total": 0.18727405749736947,
|
| 1623 |
+
"distill": 0.0004120995005731411,
|
| 1624 |
+
"task": 1.2957600558057745,
|
| 1625 |
+
"count": 47
|
| 1626 |
+
},
|
| 1627 |
+
"paraphrase": {
|
| 1628 |
+
"total": 0.016873036976903677,
|
| 1629 |
+
"distill": 0.0006976060452871024,
|
| 1630 |
+
"task": 0.23095046430826188,
|
| 1631 |
+
"count": 10
|
| 1632 |
+
}
|
| 1633 |
+
},
|
| 1634 |
+
"distill_weight": 0.2778
|
| 1635 |
+
},
|
| 1636 |
+
{
|
| 1637 |
+
"epoch": 39,
|
| 1638 |
+
"eval_results": {
|
| 1639 |
+
"sts_spearman": 0.8010951080044263,
|
| 1640 |
+
"sts_pearson": 0.7744044551086896,
|
| 1641 |
+
"retrieval_recall_at_1": 0.66,
|
| 1642 |
+
"retrieval_recall_at_5": 0.926,
|
| 1643 |
+
"retrieval_recall_at_10": 0.976,
|
| 1644 |
+
"nli_accuracy": 0.5,
|
| 1645 |
+
"nli_similarity": 0.7240303158760071,
|
| 1646 |
+
"paraphrase_accuracy": 0.5,
|
| 1647 |
+
"paraphrase_f1": 0.6666666666666666,
|
| 1648 |
+
"paraphrase_similarity": 0.8622973561286926,
|
| 1649 |
+
"composite_score": 0.7950142206688798
|
| 1650 |
+
},
|
| 1651 |
+
"losses": {
|
| 1652 |
+
"sts": {
|
| 1653 |
+
"total": 0.040290426786827004,
|
| 1654 |
+
"distill": 0.0005619644627744412,
|
| 1655 |
+
"task": 0.1388165892466255,
|
| 1656 |
+
"count": 23
|
| 1657 |
+
},
|
| 1658 |
+
"retrieval": {
|
| 1659 |
+
"total": 0.06733930118857546,
|
| 1660 |
+
"distill": 0.0006235166393378948,
|
| 1661 |
+
"task": 0.3097512569833309,
|
| 1662 |
+
"count": 47
|
| 1663 |
+
},
|
| 1664 |
+
"nli": {
|
| 1665 |
+
"total": 0.18707014081325937,
|
| 1666 |
+
"distill": 0.00041061582445028297,
|
| 1667 |
+
"task": 1.29327839993416,
|
| 1668 |
+
"count": 47
|
| 1669 |
+
},
|
| 1670 |
+
"paraphrase": {
|
| 1671 |
+
"total": 0.017511206772178413,
|
| 1672 |
+
"distill": 0.0006984172330703586,
|
| 1673 |
+
"task": 0.23959056437015533,
|
| 1674 |
+
"count": 10
|
| 1675 |
+
}
|
| 1676 |
+
},
|
| 1677 |
+
"distill_weight": 0.2772
|
| 1678 |
+
},
|
| 1679 |
+
{
|
| 1680 |
+
"epoch": 40,
|
| 1681 |
+
"eval_results": {
|
| 1682 |
+
"sts_spearman": 0.8008200971384819,
|
| 1683 |
+
"sts_pearson": 0.774130829271817,
|
| 1684 |
+
"retrieval_recall_at_1": 0.666,
|
| 1685 |
+
"retrieval_recall_at_5": 0.932,
|
| 1686 |
+
"retrieval_recall_at_10": 0.974,
|
| 1687 |
+
"nli_accuracy": 0.5,
|
| 1688 |
+
"nli_similarity": 0.7223304510116577,
|
| 1689 |
+
"paraphrase_accuracy": 0.5,
|
| 1690 |
+
"paraphrase_f1": 0.6666666666666666,
|
| 1691 |
+
"paraphrase_similarity": 0.862885594367981,
|
| 1692 |
+
"composite_score": 0.7966767152359077
|
| 1693 |
+
},
|
| 1694 |
+
"losses": {
|
| 1695 |
+
"sts": {
|
| 1696 |
+
"total": 0.039851526122378265,
|
| 1697 |
+
"distill": 0.0005616477422375718,
|
| 1698 |
+
"task": 0.1371861229772153,
|
| 1699 |
+
"count": 23
|
| 1700 |
+
},
|
| 1701 |
+
"retrieval": {
|
| 1702 |
+
"total": 0.0681393450879036,
|
| 1703 |
+
"distill": 0.0006218473711843661,
|
| 1704 |
+
"task": 0.3131846848954546,
|
| 1705 |
+
"count": 47
|
| 1706 |
+
},
|
| 1707 |
+
"nli": {
|
| 1708 |
+
"total": 0.18474397950984062,
|
| 1709 |
+
"distill": 0.00041012915824738114,
|
| 1710 |
+
"task": 1.2761304099509057,
|
| 1711 |
+
"count": 47
|
| 1712 |
+
},
|
| 1713 |
+
"paraphrase": {
|
| 1714 |
+
"total": 0.01582443034276366,
|
| 1715 |
+
"distill": 0.0006965516076888889,
|
| 1716 |
+
"task": 0.2160874292254448,
|
| 1717 |
+
"count": 10
|
| 1718 |
+
}
|
| 1719 |
+
},
|
| 1720 |
+
"distill_weight": 0.2766
|
| 1721 |
+
},
|
| 1722 |
+
{
|
| 1723 |
+
"epoch": 41,
|
| 1724 |
+
"eval_results": {
|
| 1725 |
+
"sts_spearman": 0.8004397143136981,
|
| 1726 |
+
"sts_pearson": 0.7737034113801735,
|
| 1727 |
+
"retrieval_recall_at_1": 0.662,
|
| 1728 |
+
"retrieval_recall_at_5": 0.93,
|
| 1729 |
+
"retrieval_recall_at_10": 0.976,
|
| 1730 |
+
"nli_accuracy": 0.5,
|
| 1731 |
+
"nli_similarity": 0.7216349840164185,
|
| 1732 |
+
"paraphrase_accuracy": 0.5,
|
| 1733 |
+
"paraphrase_f1": 0.6666666666666666,
|
| 1734 |
+
"paraphrase_similarity": 0.8629198670387268,
|
| 1735 |
+
"composite_score": 0.7958865238235158
|
| 1736 |
+
},
|
| 1737 |
+
"losses": {
|
| 1738 |
+
"sts": {
|
| 1739 |
+
"total": 0.040168939077335854,
|
| 1740 |
+
"distill": 0.0005606161886016312,
|
| 1741 |
+
"task": 0.1381706041486367,
|
| 1742 |
+
"count": 23
|
| 1743 |
+
},
|
| 1744 |
+
"retrieval": {
|
| 1745 |
+
"total": 0.06701801083189377,
|
| 1746 |
+
"distill": 0.0006201347891003527,
|
| 1747 |
+
"task": 0.3077663652440335,
|
| 1748 |
+
"count": 47
|
| 1749 |
+
},
|
| 1750 |
+
"nli": {
|
| 1751 |
+
"total": 0.18171446183894543,
|
| 1752 |
+
"distill": 0.00040893261246581345,
|
| 1753 |
+
"task": 1.254154608604756,
|
| 1754 |
+
"count": 47
|
| 1755 |
+
},
|
| 1756 |
+
"paraphrase": {
|
| 1757 |
+
"total": 0.016577236633747817,
|
| 1758 |
+
"distill": 0.0006964895350392908,
|
| 1759 |
+
"task": 0.2263122245669365,
|
| 1760 |
+
"count": 10
|
| 1761 |
+
}
|
| 1762 |
+
},
|
| 1763 |
+
"distill_weight": 0.276
|
| 1764 |
+
},
|
| 1765 |
+
{
|
| 1766 |
+
"epoch": 42,
|
| 1767 |
+
"eval_results": {
|
| 1768 |
+
"sts_spearman": 0.799541139671189,
|
| 1769 |
+
"sts_pearson": 0.7726619557363898,
|
| 1770 |
+
"retrieval_recall_at_1": 0.664,
|
| 1771 |
+
"retrieval_recall_at_5": 0.932,
|
| 1772 |
+
"retrieval_recall_at_10": 0.976,
|
| 1773 |
+
"nli_accuracy": 0.5,
|
| 1774 |
+
"nli_similarity": 0.7218517065048218,
|
| 1775 |
+
"paraphrase_accuracy": 0.5,
|
| 1776 |
+
"paraphrase_f1": 0.6666666666666666,
|
| 1777 |
+
"paraphrase_similarity": 0.862092912197113,
|
| 1778 |
+
"composite_score": 0.7960372365022612
|
| 1779 |
+
},
|
| 1780 |
+
"losses": {
|
| 1781 |
+
"sts": {
|
| 1782 |
+
"total": 0.039756916949282524,
|
| 1783 |
+
"distill": 0.0005605145656179799,
|
| 1784 |
+
"task": 0.13663589921982391,
|
| 1785 |
+
"count": 23
|
| 1786 |
+
},
|
| 1787 |
+
"retrieval": {
|
| 1788 |
+
"total": 0.06787618566700752,
|
| 1789 |
+
"distill": 0.0006185190620871776,
|
| 1790 |
+
"task": 0.3114630852607971,
|
| 1791 |
+
"count": 47
|
| 1792 |
+
},
|
| 1793 |
+
"nli": {
|
| 1794 |
+
"total": 0.1818689741986863,
|
| 1795 |
+
"distill": 0.0004076707292041008,
|
| 1796 |
+
"task": 1.254186414657755,
|
| 1797 |
+
"count": 47
|
| 1798 |
+
},
|
| 1799 |
+
"paraphrase": {
|
| 1800 |
+
"total": 0.015548030100762843,
|
| 1801 |
+
"distill": 0.0006947330140974373,
|
| 1802 |
+
"task": 0.21193347945809365,
|
| 1803 |
+
"count": 10
|
| 1804 |
+
}
|
| 1805 |
+
},
|
| 1806 |
+
"distill_weight": 0.2754
|
| 1807 |
+
},
|
| 1808 |
+
{
|
| 1809 |
+
"epoch": 43,
|
| 1810 |
+
"eval_results": {
|
| 1811 |
+
"sts_spearman": 0.798959430373989,
|
| 1812 |
+
"sts_pearson": 0.7717844926047425,
|
| 1813 |
+
"retrieval_recall_at_1": 0.658,
|
| 1814 |
+
"retrieval_recall_at_5": 0.934,
|
| 1815 |
+
"retrieval_recall_at_10": 0.976,
|
| 1816 |
+
"nli_accuracy": 0.5,
|
| 1817 |
+
"nli_similarity": 0.7230818867683411,
|
| 1818 |
+
"paraphrase_accuracy": 0.5,
|
| 1819 |
+
"paraphrase_f1": 0.6666666666666666,
|
| 1820 |
+
"paraphrase_similarity": 0.8633172512054443,
|
| 1821 |
+
"composite_score": 0.7963463818536611
|
| 1822 |
+
},
|
| 1823 |
+
"losses": {
|
| 1824 |
+
"sts": {
|
| 1825 |
+
"total": 0.03980643334596053,
|
| 1826 |
+
"distill": 0.0005611540931884361,
|
| 1827 |
+
"task": 0.13669410898633624,
|
| 1828 |
+
"count": 23
|
| 1829 |
+
},
|
| 1830 |
+
"retrieval": {
|
| 1831 |
+
"total": 0.0670052819905129,
|
| 1832 |
+
"distill": 0.0006171439581134535,
|
| 1833 |
+
"task": 0.3072057971928982,
|
| 1834 |
+
"count": 47
|
| 1835 |
+
},
|
| 1836 |
+
"nli": {
|
| 1837 |
+
"total": 0.181170218168421,
|
| 1838 |
+
"distill": 0.0004060288384665755,
|
| 1839 |
+
"task": 1.2483358712906534,
|
| 1840 |
+
"count": 47
|
| 1841 |
+
},
|
| 1842 |
+
"paraphrase": {
|
| 1843 |
+
"total": 0.016034624353051186,
|
| 1844 |
+
"distill": 0.0006926223228219897,
|
| 1845 |
+
"task": 0.21848167926073075,
|
| 1846 |
+
"count": 10
|
| 1847 |
+
}
|
| 1848 |
+
},
|
| 1849 |
+
"distill_weight": 0.2748
|
| 1850 |
+
},
|
| 1851 |
+
{
|
| 1852 |
+
"epoch": 44,
|
| 1853 |
+
"eval_results": {
|
| 1854 |
+
"sts_spearman": 0.7992766546129851,
|
| 1855 |
+
"sts_pearson": 0.7723631373159269,
|
| 1856 |
+
"retrieval_recall_at_1": 0.668,
|
| 1857 |
+
"retrieval_recall_at_5": 0.936,
|
| 1858 |
+
"retrieval_recall_at_10": 0.976,
|
| 1859 |
+
"nli_accuracy": 0.5,
|
| 1860 |
+
"nli_similarity": 0.7214555740356445,
|
| 1861 |
+
"paraphrase_accuracy": 0.5,
|
| 1862 |
+
"paraphrase_f1": 0.6666666666666666,
|
| 1863 |
+
"paraphrase_similarity": 0.8633242845535278,
|
| 1864 |
+
"composite_score": 0.7971049939731593
|
| 1865 |
+
},
|
| 1866 |
+
"losses": {
|
| 1867 |
+
"sts": {
|
| 1868 |
+
"total": 0.03956834745147954,
|
| 1869 |
+
"distill": 0.0005598463517937647,
|
| 1870 |
+
"task": 0.13576342230257782,
|
| 1871 |
+
"count": 23
|
| 1872 |
+
},
|
| 1873 |
+
"retrieval": {
|
| 1874 |
+
"total": 0.0675140564587522,
|
| 1875 |
+
"distill": 0.0006155266111934597,
|
| 1876 |
+
"task": 0.30929218010699494,
|
| 1877 |
+
"count": 47
|
| 1878 |
+
},
|
| 1879 |
+
"nli": {
|
| 1880 |
+
"total": 0.18059916984527669,
|
| 1881 |
+
"distill": 0.00040597010755594426,
|
| 1882 |
+
"task": 1.2433717808824905,
|
| 1883 |
+
"count": 47
|
| 1884 |
+
},
|
| 1885 |
+
"paraphrase": {
|
| 1886 |
+
"total": 0.016109541151672603,
|
| 1887 |
+
"distill": 0.000693465251242742,
|
| 1888 |
+
"task": 0.21933580189943314,
|
| 1889 |
+
"count": 10
|
| 1890 |
+
}
|
| 1891 |
+
},
|
| 1892 |
+
"distill_weight": 0.2742
|
| 1893 |
+
},
|
| 1894 |
+
{
|
| 1895 |
+
"epoch": 45,
|
| 1896 |
+
"eval_results": {
|
| 1897 |
+
"sts_spearman": 0.7987902718942615,
|
| 1898 |
+
"sts_pearson": 0.7716241234848017,
|
| 1899 |
+
"retrieval_recall_at_1": 0.67,
|
| 1900 |
+
"retrieval_recall_at_5": 0.934,
|
| 1901 |
+
"retrieval_recall_at_10": 0.976,
|
| 1902 |
+
"nli_accuracy": 0.5,
|
| 1903 |
+
"nli_similarity": 0.7223896384239197,
|
| 1904 |
+
"paraphrase_accuracy": 0.5,
|
| 1905 |
+
"paraphrase_f1": 0.6666666666666666,
|
| 1906 |
+
"paraphrase_similarity": 0.8632831573486328,
|
| 1907 |
+
"composite_score": 0.7962618026137975
|
| 1908 |
+
},
|
| 1909 |
+
"losses": {
|
| 1910 |
+
"sts": {
|
| 1911 |
+
"total": 0.03940606570762137,
|
| 1912 |
+
"distill": 0.0005587598197567074,
|
| 1913 |
+
"task": 0.13509494876084122,
|
| 1914 |
+
"count": 23
|
| 1915 |
+
},
|
| 1916 |
+
"retrieval": {
|
| 1917 |
+
"total": 0.06665098445212587,
|
| 1918 |
+
"distill": 0.0006136801812124062,
|
| 1919 |
+
"task": 0.3050802118600683,
|
| 1920 |
+
"count": 47
|
| 1921 |
+
},
|
| 1922 |
+
"nli": {
|
| 1923 |
+
"total": 0.18052284007376812,
|
| 1924 |
+
"distill": 0.00040472789662592906,
|
| 1925 |
+
"task": 1.2418234031251136,
|
| 1926 |
+
"count": 47
|
| 1927 |
+
},
|
| 1928 |
+
"paraphrase": {
|
| 1929 |
+
"total": 0.015592311229556798,
|
| 1930 |
+
"distill": 0.0006917931721545755,
|
| 1931 |
+
"task": 0.21204619854688644,
|
| 1932 |
+
"count": 10
|
| 1933 |
+
}
|
| 1934 |
+
},
|
| 1935 |
+
"distill_weight": 0.2736
|
| 1936 |
+
},
|
| 1937 |
+
{
|
| 1938 |
+
"epoch": 46,
|
| 1939 |
+
"eval_results": {
|
| 1940 |
+
"sts_spearman": 0.7980845899126008,
|
| 1941 |
+
"sts_pearson": 0.7708174356684444,
|
| 1942 |
+
"retrieval_recall_at_1": 0.662,
|
| 1943 |
+
"retrieval_recall_at_5": 0.934,
|
| 1944 |
+
"retrieval_recall_at_10": 0.974,
|
| 1945 |
+
"nli_accuracy": 0.5,
|
| 1946 |
+
"nli_similarity": 0.7227649688720703,
|
| 1947 |
+
"paraphrase_accuracy": 0.5,
|
| 1948 |
+
"paraphrase_f1": 0.6666666666666666,
|
| 1949 |
+
"paraphrase_similarity": 0.8638308644294739,
|
| 1950 |
+
"composite_score": 0.7959089616229671
|
| 1951 |
+
},
|
| 1952 |
+
"losses": {
|
| 1953 |
+
"sts": {
|
| 1954 |
+
"total": 0.03914242918076723,
|
| 1955 |
+
"distill": 0.0005580810273228133,
|
| 1956 |
+
"task": 0.13407865684965384,
|
| 1957 |
+
"count": 23
|
| 1958 |
+
},
|
| 1959 |
+
"retrieval": {
|
| 1960 |
+
"total": 0.06371399538314089,
|
| 1961 |
+
"distill": 0.0006122274517497801,
|
| 1962 |
+
"task": 0.2913656944924213,
|
| 1963 |
+
"count": 47
|
| 1964 |
+
},
|
| 1965 |
+
"nli": {
|
| 1966 |
+
"total": 0.1788206826499168,
|
| 1967 |
+
"distill": 0.00040350851176821804,
|
| 1968 |
+
"task": 1.2290957481303113,
|
| 1969 |
+
"count": 47
|
| 1970 |
+
},
|
| 1971 |
+
"paraphrase": {
|
| 1972 |
+
"total": 0.016117287054657935,
|
| 1973 |
+
"distill": 0.0006897407933138311,
|
| 1974 |
+
"task": 0.2191057413816452,
|
| 1975 |
+
"count": 10
|
| 1976 |
+
}
|
| 1977 |
+
},
|
| 1978 |
+
"distill_weight": 0.273
|
| 1979 |
+
},
|
| 1980 |
+
{
|
| 1981 |
+
"epoch": 47,
|
| 1982 |
+
"eval_results": {
|
| 1983 |
+
"sts_spearman": 0.7986050030831314,
|
| 1984 |
+
"sts_pearson": 0.7713344721815415,
|
| 1985 |
+
"retrieval_recall_at_1": 0.668,
|
| 1986 |
+
"retrieval_recall_at_5": 0.936,
|
| 1987 |
+
"retrieval_recall_at_10": 0.974,
|
| 1988 |
+
"nli_accuracy": 0.5,
|
| 1989 |
+
"nli_similarity": 0.7211058139801025,
|
| 1990 |
+
"paraphrase_accuracy": 0.5,
|
| 1991 |
+
"paraphrase_f1": 0.6666666666666666,
|
| 1992 |
+
"paraphrase_similarity": 0.8641474843025208,
|
| 1993 |
+
"composite_score": 0.7967691682082324
|
| 1994 |
+
},
|
| 1995 |
+
"losses": {
|
| 1996 |
+
"sts": {
|
| 1997 |
+
"total": 0.03924534291676853,
|
| 1998 |
+
"distill": 0.0005579624380713896,
|
| 1999 |
+
"task": 0.1343229581480441,
|
| 2000 |
+
"count": 23
|
| 2001 |
+
},
|
| 2002 |
+
"retrieval": {
|
| 2003 |
+
"total": 0.06390803307294846,
|
| 2004 |
+
"distill": 0.0006106501616081817,
|
| 2005 |
+
"task": 0.29201801152939494,
|
| 2006 |
+
"count": 47
|
| 2007 |
+
},
|
| 2008 |
+
"nli": {
|
| 2009 |
+
"total": 0.17669346231095334,
|
| 2010 |
+
"distill": 0.0004022118878213966,
|
| 2011 |
+
"task": 1.2134682447352307,
|
| 2012 |
+
"count": 47
|
| 2013 |
+
},
|
| 2014 |
+
"paraphrase": {
|
| 2015 |
+
"total": 0.014580465480685234,
|
| 2016 |
+
"distill": 0.0006913242454174906,
|
| 2017 |
+
"task": 0.1978030323982239,
|
| 2018 |
+
"count": 10
|
| 2019 |
+
}
|
| 2020 |
+
},
|
| 2021 |
+
"distill_weight": 0.2724
|
| 2022 |
+
},
|
| 2023 |
+
{
|
| 2024 |
+
"epoch": 48,
|
| 2025 |
+
"eval_results": {
|
| 2026 |
+
"sts_spearman": 0.7978328086000539,
|
| 2027 |
+
"sts_pearson": 0.7705122749494036,
|
| 2028 |
+
"retrieval_recall_at_1": 0.67,
|
| 2029 |
+
"retrieval_recall_at_5": 0.936,
|
| 2030 |
+
"retrieval_recall_at_10": 0.976,
|
| 2031 |
+
"nli_accuracy": 0.5,
|
| 2032 |
+
"nli_similarity": 0.7213853001594543,
|
| 2033 |
+
"paraphrase_accuracy": 0.5,
|
| 2034 |
+
"paraphrase_f1": 0.6666666666666666,
|
| 2035 |
+
"paraphrase_similarity": 0.8641984462738037,
|
| 2036 |
+
"composite_score": 0.7963830709666937
|
| 2037 |
+
},
|
| 2038 |
+
"losses": {
|
| 2039 |
+
"sts": {
|
| 2040 |
+
"total": 0.038881031796336174,
|
| 2041 |
+
"distill": 0.0005576102867843988,
|
| 2042 |
+
"task": 0.13296303638945456,
|
| 2043 |
+
"count": 23
|
| 2044 |
+
},
|
| 2045 |
+
"retrieval": {
|
| 2046 |
+
"total": 0.06267972861198669,
|
| 2047 |
+
"distill": 0.0006095131591675764,
|
| 2048 |
+
"task": 0.2861579453691523,
|
| 2049 |
+
"count": 47
|
| 2050 |
+
},
|
| 2051 |
+
"nli": {
|
| 2052 |
+
"total": 0.17544851499669095,
|
| 2053 |
+
"distill": 0.0004012782276518881,
|
| 2054 |
+
"task": 1.2039236971672544,
|
| 2055 |
+
"count": 47
|
| 2056 |
+
},
|
| 2057 |
+
"paraphrase": {
|
| 2058 |
+
"total": 0.015775612369179726,
|
| 2059 |
+
"distill": 0.0006886643706820906,
|
| 2060 |
+
"task": 0.2140680193901062,
|
| 2061 |
+
"count": 10
|
| 2062 |
+
}
|
| 2063 |
+
},
|
| 2064 |
+
"distill_weight": 0.2718
|
| 2065 |
+
},
|
| 2066 |
+
{
|
| 2067 |
+
"epoch": 49,
|
| 2068 |
+
"eval_results": {
|
| 2069 |
+
"sts_spearman": 0.7973172210556293,
|
| 2070 |
+
"sts_pearson": 0.7701713894839444,
|
| 2071 |
+
"retrieval_recall_at_1": 0.67,
|
| 2072 |
+
"retrieval_recall_at_5": 0.936,
|
| 2073 |
+
"retrieval_recall_at_10": 0.976,
|
| 2074 |
+
"nli_accuracy": 0.5,
|
| 2075 |
+
"nli_similarity": 0.7203776240348816,
|
| 2076 |
+
"paraphrase_accuracy": 0.5,
|
| 2077 |
+
"paraphrase_f1": 0.6666666666666666,
|
| 2078 |
+
"paraphrase_similarity": 0.8631545305252075,
|
| 2079 |
+
"composite_score": 0.7961252771944813
|
| 2080 |
+
},
|
| 2081 |
+
"losses": {
|
| 2082 |
+
"sts": {
|
| 2083 |
+
"total": 0.03882810230488363,
|
| 2084 |
+
"distill": 0.0005557498307493718,
|
| 2085 |
+
"task": 0.13267488712849823,
|
| 2086 |
+
"count": 23
|
| 2087 |
+
},
|
| 2088 |
+
"retrieval": {
|
| 2089 |
+
"total": 0.0626571145146451,
|
| 2090 |
+
"distill": 0.000607952565955434,
|
| 2091 |
+
"task": 0.28582252847387435,
|
| 2092 |
+
"count": 47
|
| 2093 |
+
},
|
| 2094 |
+
"nli": {
|
| 2095 |
+
"total": 0.1763826126747943,
|
| 2096 |
+
"distill": 0.0004007951009701541,
|
| 2097 |
+
"task": 1.2093435627348879,
|
| 2098 |
+
"count": 47
|
| 2099 |
+
},
|
| 2100 |
+
"paraphrase": {
|
| 2101 |
+
"total": 0.015177728608250618,
|
| 2102 |
+
"distill": 0.0006882219109684229,
|
| 2103 |
+
"task": 0.20569542646408082,
|
| 2104 |
+
"count": 10
|
| 2105 |
+
}
|
| 2106 |
+
},
|
| 2107 |
+
"distill_weight": 0.2712
|
| 2108 |
+
},
|
| 2109 |
+
{
|
| 2110 |
+
"epoch": 50,
|
| 2111 |
+
"eval_results": {
|
| 2112 |
+
"sts_spearman": 0.7968594218073127,
|
| 2113 |
+
"sts_pearson": 0.7695483662129042,
|
| 2114 |
+
"retrieval_recall_at_1": 0.67,
|
| 2115 |
+
"retrieval_recall_at_5": 0.936,
|
| 2116 |
+
"retrieval_recall_at_10": 0.974,
|
| 2117 |
+
"nli_accuracy": 0.5,
|
| 2118 |
+
"nli_similarity": 0.7204389572143555,
|
| 2119 |
+
"paraphrase_accuracy": 0.5,
|
| 2120 |
+
"paraphrase_f1": 0.6666666666666666,
|
| 2121 |
+
"paraphrase_similarity": 0.8636235594749451,
|
| 2122 |
+
"composite_score": 0.795896377570323
|
| 2123 |
+
},
|
| 2124 |
+
"losses": {
|
| 2125 |
+
"sts": {
|
| 2126 |
+
"total": 0.03918402480042499,
|
| 2127 |
+
"distill": 0.0005560298273137406,
|
| 2128 |
+
"task": 0.13378654711920282,
|
| 2129 |
+
"count": 23
|
| 2130 |
+
},
|
| 2131 |
+
"retrieval": {
|
| 2132 |
+
"total": 0.06285070390143294,
|
| 2133 |
+
"distill": 0.000605922420211929,
|
| 2134 |
+
"task": 0.2864762779245985,
|
| 2135 |
+
"count": 47
|
| 2136 |
+
},
|
| 2137 |
+
"nli": {
|
| 2138 |
+
"total": 0.1742387483728693,
|
| 2139 |
+
"distill": 0.00039892817941553733,
|
| 2140 |
+
"task": 1.1936577951654475,
|
| 2141 |
+
"count": 47
|
| 2142 |
+
},
|
| 2143 |
+
"paraphrase": {
|
| 2144 |
+
"total": 0.014562566205859185,
|
| 2145 |
+
"distill": 0.0006869392120279372,
|
| 2146 |
+
"task": 0.197102826833725,
|
| 2147 |
+
"count": 10
|
| 2148 |
+
}
|
| 2149 |
+
},
|
| 2150 |
+
"distill_weight": 0.2706
|
| 2151 |
+
},
|
| 2152 |
+
{
|
| 2153 |
+
"epoch": 51,
|
| 2154 |
+
"eval_results": {
|
| 2155 |
+
"sts_spearman": 0.7961018874767011,
|
| 2156 |
+
"sts_pearson": 0.7685896198859151,
|
| 2157 |
+
"retrieval_recall_at_1": 0.672,
|
| 2158 |
+
"retrieval_recall_at_5": 0.938,
|
| 2159 |
+
"retrieval_recall_at_10": 0.976,
|
| 2160 |
+
"nli_accuracy": 0.5,
|
| 2161 |
+
"nli_similarity": 0.7212554216384888,
|
| 2162 |
+
"paraphrase_accuracy": 0.5,
|
| 2163 |
+
"paraphrase_f1": 0.6666666666666666,
|
| 2164 |
+
"paraphrase_similarity": 0.8639085292816162,
|
| 2165 |
+
"composite_score": 0.7961176104050172
|
| 2166 |
+
},
|
| 2167 |
+
"losses": {
|
| 2168 |
+
"sts": {
|
| 2169 |
+
"total": 0.038832955305343086,
|
| 2170 |
+
"distill": 0.000555693042849231,
|
| 2171 |
+
"task": 0.13247574898211853,
|
| 2172 |
+
"count": 23
|
| 2173 |
+
},
|
| 2174 |
+
"retrieval": {
|
| 2175 |
+
"total": 0.061906919121108156,
|
| 2176 |
+
"distill": 0.0006047165407700107,
|
| 2177 |
+
"task": 0.2819344585246228,
|
| 2178 |
+
"count": 47
|
| 2179 |
+
},
|
| 2180 |
+
"nli": {
|
| 2181 |
+
"total": 0.17475084611710082,
|
| 2182 |
+
"distill": 0.0003985753827827408,
|
| 2183 |
+
"task": 1.1961865387064345,
|
| 2184 |
+
"count": 47
|
| 2185 |
+
},
|
| 2186 |
+
"paraphrase": {
|
| 2187 |
+
"total": 0.014561035577207804,
|
| 2188 |
+
"distill": 0.0006853513652458787,
|
| 2189 |
+
"task": 0.19693138003349303,
|
| 2190 |
+
"count": 10
|
| 2191 |
+
}
|
| 2192 |
+
},
|
| 2193 |
+
"distill_weight": 0.27
|
| 2194 |
+
},
|
| 2195 |
+
{
|
| 2196 |
+
"epoch": 52,
|
| 2197 |
+
"eval_results": {
|
| 2198 |
+
"sts_spearman": 0.7957712004157503,
|
| 2199 |
+
"sts_pearson": 0.7682444812474918,
|
| 2200 |
+
"retrieval_recall_at_1": 0.668,
|
| 2201 |
+
"retrieval_recall_at_5": 0.938,
|
| 2202 |
+
"retrieval_recall_at_10": 0.976,
|
| 2203 |
+
"nli_accuracy": 0.5,
|
| 2204 |
+
"nli_similarity": 0.7211222052574158,
|
| 2205 |
+
"paraphrase_accuracy": 0.5,
|
| 2206 |
+
"paraphrase_f1": 0.6666666666666666,
|
| 2207 |
+
"paraphrase_similarity": 0.8639959096908569,
|
| 2208 |
+
"composite_score": 0.7959522668745418
|
| 2209 |
+
},
|
| 2210 |
+
"losses": {
|
| 2211 |
+
"sts": {
|
| 2212 |
+
"total": 0.038104739849982056,
|
| 2213 |
+
"distill": 0.0005552242105097874,
|
| 2214 |
+
"task": 0.12987668682699618,
|
| 2215 |
+
"count": 23
|
| 2216 |
+
},
|
| 2217 |
+
"retrieval": {
|
| 2218 |
+
"total": 0.06224994496145147,
|
| 2219 |
+
"distill": 0.0006037014598482625,
|
| 2220 |
+
"task": 0.2832708587037756,
|
| 2221 |
+
"count": 47
|
| 2222 |
+
},
|
| 2223 |
+
"nli": {
|
| 2224 |
+
"total": 0.17524569560872746,
|
| 2225 |
+
"distill": 0.00039716928756419333,
|
| 2226 |
+
"task": 1.198594996269713,
|
| 2227 |
+
"count": 47
|
| 2228 |
+
},
|
| 2229 |
+
"paraphrase": {
|
| 2230 |
+
"total": 0.01391328014433384,
|
| 2231 |
+
"distill": 0.0006847139564342796,
|
| 2232 |
+
"task": 0.18791155964136125,
|
| 2233 |
+
"count": 10
|
| 2234 |
+
}
|
| 2235 |
+
},
|
| 2236 |
+
"distill_weight": 0.2694
|
| 2237 |
+
},
|
| 2238 |
+
{
|
| 2239 |
+
"epoch": 53,
|
| 2240 |
+
"eval_results": {
|
| 2241 |
+
"sts_spearman": 0.7948627602870467,
|
| 2242 |
+
"sts_pearson": 0.7672766924578068,
|
| 2243 |
+
"retrieval_recall_at_1": 0.674,
|
| 2244 |
+
"retrieval_recall_at_5": 0.936,
|
| 2245 |
+
"retrieval_recall_at_10": 0.974,
|
| 2246 |
+
"nli_accuracy": 0.5,
|
| 2247 |
+
"nli_similarity": 0.7208473682403564,
|
| 2248 |
+
"paraphrase_accuracy": 0.5,
|
| 2249 |
+
"paraphrase_f1": 0.6666666666666666,
|
| 2250 |
+
"paraphrase_similarity": 0.8635191321372986,
|
| 2251 |
+
"composite_score": 0.79489804681019
|
| 2252 |
+
},
|
| 2253 |
+
"losses": {
|
| 2254 |
+
"sts": {
|
| 2255 |
+
"total": 0.03807506864161595,
|
| 2256 |
+
"distill": 0.0005540588349306389,
|
| 2257 |
+
"task": 0.12967087879129077,
|
| 2258 |
+
"count": 23
|
| 2259 |
+
},
|
| 2260 |
+
"retrieval": {
|
| 2261 |
+
"total": 0.06309370720323096,
|
| 2262 |
+
"distill": 0.0006028104070673122,
|
| 2263 |
+
"task": 0.2868876425509757,
|
| 2264 |
+
"count": 47
|
| 2265 |
+
},
|
| 2266 |
+
"nli": {
|
| 2267 |
+
"total": 0.17431603213574023,
|
| 2268 |
+
"distill": 0.00039660765918249146,
|
| 2269 |
+
"task": 1.1912570329422647,
|
| 2270 |
+
"count": 47
|
| 2271 |
+
},
|
| 2272 |
+
"paraphrase": {
|
| 2273 |
+
"total": 0.013855330273509025,
|
| 2274 |
+
"distill": 0.0006855725019704551,
|
| 2275 |
+
"task": 0.18696729764342307,
|
| 2276 |
+
"count": 10
|
| 2277 |
+
}
|
| 2278 |
+
},
|
| 2279 |
+
"distill_weight": 0.2688
|
| 2280 |
+
},
|
| 2281 |
+
{
|
| 2282 |
+
"epoch": 54,
|
| 2283 |
+
"eval_results": {
|
| 2284 |
+
"sts_spearman": 0.7948670645604796,
|
| 2285 |
+
"sts_pearson": 0.7673114563945611,
|
| 2286 |
+
"retrieval_recall_at_1": 0.672,
|
| 2287 |
+
"retrieval_recall_at_5": 0.936,
|
| 2288 |
+
"retrieval_recall_at_10": 0.976,
|
| 2289 |
+
"nli_accuracy": 0.5,
|
| 2290 |
+
"nli_similarity": 0.7204586267471313,
|
| 2291 |
+
"paraphrase_accuracy": 0.5,
|
| 2292 |
+
"paraphrase_f1": 0.6666666666666666,
|
| 2293 |
+
"paraphrase_similarity": 0.8637556433677673,
|
| 2294 |
+
"composite_score": 0.7949001989469064
|
| 2295 |
+
},
|
| 2296 |
+
"losses": {
|
| 2297 |
+
"sts": {
|
| 2298 |
+
"total": 0.03859076422193776,
|
| 2299 |
+
"distill": 0.000553890953913493,
|
| 2300 |
+
"task": 0.1313275915125142,
|
| 2301 |
+
"count": 23
|
| 2302 |
+
},
|
| 2303 |
+
"retrieval": {
|
| 2304 |
+
"total": 0.06243176108345072,
|
| 2305 |
+
"distill": 0.000601062821125255,
|
| 2306 |
+
"task": 0.2836410492024523,
|
| 2307 |
+
"count": 47
|
| 2308 |
+
},
|
| 2309 |
+
"nli": {
|
| 2310 |
+
"total": 0.17191652573169547,
|
| 2311 |
+
"distill": 0.0003955346752612039,
|
| 2312 |
+
"task": 1.173889383356622,
|
| 2313 |
+
"count": 47
|
| 2314 |
+
},
|
| 2315 |
+
"paraphrase": {
|
| 2316 |
+
"total": 0.014081115927547216,
|
| 2317 |
+
"distill": 0.0006832460989244282,
|
| 2318 |
+
"task": 0.1899135023355484,
|
| 2319 |
+
"count": 10
|
| 2320 |
+
}
|
| 2321 |
+
},
|
| 2322 |
+
"distill_weight": 0.2682
|
| 2323 |
+
},
|
| 2324 |
+
{
|
| 2325 |
+
"epoch": 55,
|
| 2326 |
+
"eval_results": {
|
| 2327 |
+
"sts_spearman": 0.7940184198467072,
|
| 2328 |
+
"sts_pearson": 0.7664062690593959,
|
| 2329 |
+
"retrieval_recall_at_1": 0.67,
|
| 2330 |
+
"retrieval_recall_at_5": 0.936,
|
| 2331 |
+
"retrieval_recall_at_10": 0.974,
|
| 2332 |
+
"nli_accuracy": 0.5,
|
| 2333 |
+
"nli_similarity": 0.7194271087646484,
|
| 2334 |
+
"paraphrase_accuracy": 0.5,
|
| 2335 |
+
"paraphrase_f1": 0.6666666666666666,
|
| 2336 |
+
"paraphrase_similarity": 0.8639047145843506,
|
| 2337 |
+
"composite_score": 0.7944758765900203
|
| 2338 |
+
},
|
| 2339 |
+
"losses": {
|
| 2340 |
+
"sts": {
|
| 2341 |
+
"total": 0.03801367480469787,
|
| 2342 |
+
"distill": 0.0005527152021860946,
|
| 2343 |
+
"task": 0.12925235538379007,
|
| 2344 |
+
"count": 23
|
| 2345 |
+
},
|
| 2346 |
+
"retrieval": {
|
| 2347 |
+
"total": 0.06092823788206628,
|
| 2348 |
+
"distill": 0.0005999853217756336,
|
| 2349 |
+
"task": 0.276568721900595,
|
| 2350 |
+
"count": 47
|
| 2351 |
+
},
|
| 2352 |
+
"nli": {
|
| 2353 |
+
"total": 0.1702281376148792,
|
| 2354 |
+
"distill": 0.0003946685254514376,
|
| 2355 |
+
"task": 1.161404493007254,
|
| 2356 |
+
"count": 47
|
| 2357 |
+
},
|
| 2358 |
+
"paraphrase": {
|
| 2359 |
+
"total": 0.012735517043620349,
|
| 2360 |
+
"distill": 0.0006815990433096886,
|
| 2361 |
+
"task": 0.17139707654714584,
|
| 2362 |
+
"count": 10
|
| 2363 |
+
}
|
| 2364 |
+
},
|
| 2365 |
+
"distill_weight": 0.2676
|
| 2366 |
+
},
|
| 2367 |
+
{
|
| 2368 |
+
"epoch": 56,
|
| 2369 |
+
"eval_results": {
|
| 2370 |
+
"sts_spearman": 0.7940615711220378,
|
| 2371 |
+
"sts_pearson": 0.7665661594709636,
|
| 2372 |
+
"retrieval_recall_at_1": 0.672,
|
| 2373 |
+
"retrieval_recall_at_5": 0.94,
|
| 2374 |
+
"retrieval_recall_at_10": 0.976,
|
| 2375 |
+
"nli_accuracy": 0.5,
|
| 2376 |
+
"nli_similarity": 0.7197932004928589,
|
| 2377 |
+
"paraphrase_accuracy": 0.5,
|
| 2378 |
+
"paraphrase_f1": 0.6666666666666666,
|
| 2379 |
+
"paraphrase_similarity": 0.8638368248939514,
|
| 2380 |
+
"composite_score": 0.7956974522276856
|
| 2381 |
+
},
|
| 2382 |
+
"losses": {
|
| 2383 |
+
"sts": {
|
| 2384 |
+
"total": 0.03830795994271403,
|
| 2385 |
+
"distill": 0.0005517558893188834,
|
| 2386 |
+
"task": 0.13015226015578146,
|
| 2387 |
+
"count": 23
|
| 2388 |
+
},
|
| 2389 |
+
"retrieval": {
|
| 2390 |
+
"total": 0.060107594633356055,
|
| 2391 |
+
"distill": 0.0005995195837037519,
|
| 2392 |
+
"task": 0.27261265953804587,
|
| 2393 |
+
"count": 47
|
| 2394 |
+
},
|
| 2395 |
+
"nli": {
|
| 2396 |
+
"total": 0.1709233709472291,
|
| 2397 |
+
"distill": 0.0003937529740567775,
|
| 2398 |
+
"task": 1.165199505521896,
|
| 2399 |
+
"count": 47
|
| 2400 |
+
},
|
| 2401 |
+
"paraphrase": {
|
| 2402 |
+
"total": 0.014191682077944278,
|
| 2403 |
+
"distill": 0.0006824985903222114,
|
| 2404 |
+
"task": 0.19112490341067315,
|
| 2405 |
+
"count": 10
|
| 2406 |
+
}
|
| 2407 |
+
},
|
| 2408 |
+
"distill_weight": 0.267
|
| 2409 |
+
},
|
| 2410 |
+
{
|
| 2411 |
+
"epoch": 57,
|
| 2412 |
+
"eval_results": {
|
| 2413 |
+
"sts_spearman": 0.7937036029034699,
|
| 2414 |
+
"sts_pearson": 0.7662776909565029,
|
| 2415 |
+
"retrieval_recall_at_1": 0.672,
|
| 2416 |
+
"retrieval_recall_at_5": 0.942,
|
| 2417 |
+
"retrieval_recall_at_10": 0.978,
|
| 2418 |
+
"nli_accuracy": 0.5,
|
| 2419 |
+
"nli_similarity": 0.7196366190910339,
|
| 2420 |
+
"paraphrase_accuracy": 0.5,
|
| 2421 |
+
"paraphrase_f1": 0.6666666666666666,
|
| 2422 |
+
"paraphrase_similarity": 0.864041268825531,
|
| 2423 |
+
"composite_score": 0.7961184681184016
|
| 2424 |
+
},
|
| 2425 |
+
"losses": {
|
| 2426 |
+
"sts": {
|
| 2427 |
+
"total": 0.038414100913897804,
|
| 2428 |
+
"distill": 0.0005509398837128411,
|
| 2429 |
+
"task": 0.1304093819597493,
|
| 2430 |
+
"count": 23
|
| 2431 |
+
},
|
| 2432 |
+
"retrieval": {
|
| 2433 |
+
"total": 0.057978192938769116,
|
| 2434 |
+
"distill": 0.0005975660850650611,
|
| 2435 |
+
"task": 0.26271810398456896,
|
| 2436 |
+
"count": 47
|
| 2437 |
+
},
|
| 2438 |
+
"nli": {
|
| 2439 |
+
"total": 0.17013170490873622,
|
| 2440 |
+
"distill": 0.00039284772232194687,
|
| 2441 |
+
"task": 1.1588538897798417,
|
| 2442 |
+
"count": 47
|
| 2443 |
+
},
|
| 2444 |
+
"paraphrase": {
|
| 2445 |
+
"total": 0.014643667824566364,
|
| 2446 |
+
"distill": 0.0006814012362156063,
|
| 2447 |
+
"task": 0.19713934063911437,
|
| 2448 |
+
"count": 10
|
| 2449 |
+
}
|
| 2450 |
+
},
|
| 2451 |
+
"distill_weight": 0.26639999999999997
|
| 2452 |
+
},
|
| 2453 |
+
{
|
| 2454 |
+
"epoch": 58,
|
| 2455 |
+
"eval_results": {
|
| 2456 |
+
"sts_spearman": 0.7928266663552261,
|
| 2457 |
+
"sts_pearson": 0.765244563485406,
|
| 2458 |
+
"retrieval_recall_at_1": 0.678,
|
| 2459 |
+
"retrieval_recall_at_5": 0.94,
|
| 2460 |
+
"retrieval_recall_at_10": 0.976,
|
| 2461 |
+
"nli_accuracy": 0.5,
|
| 2462 |
+
"nli_similarity": 0.7203112840652466,
|
| 2463 |
+
"paraphrase_accuracy": 0.5,
|
| 2464 |
+
"paraphrase_f1": 0.6666666666666666,
|
| 2465 |
+
"paraphrase_similarity": 0.8644272685050964,
|
| 2466 |
+
"composite_score": 0.7950799998442797
|
| 2467 |
+
},
|
| 2468 |
+
"losses": {
|
| 2469 |
+
"sts": {
|
| 2470 |
+
"total": 0.03789730612998423,
|
| 2471 |
+
"distill": 0.0005501627937242713,
|
| 2472 |
+
"task": 0.1285449188688527,
|
| 2473 |
+
"count": 23
|
| 2474 |
+
},
|
| 2475 |
+
"retrieval": {
|
| 2476 |
+
"total": 0.06108237723720834,
|
| 2477 |
+
"distill": 0.0005970121710561216,
|
| 2478 |
+
"task": 0.27659898869534755,
|
| 2479 |
+
"count": 47
|
| 2480 |
+
},
|
| 2481 |
+
"nli": {
|
| 2482 |
+
"total": 0.1709375616083754,
|
| 2483 |
+
"distill": 0.0003915692932230044,
|
| 2484 |
+
"task": 1.1633987756485635,
|
| 2485 |
+
"count": 47
|
| 2486 |
+
},
|
| 2487 |
+
"paraphrase": {
|
| 2488 |
+
"total": 0.013374552130699158,
|
| 2489 |
+
"distill": 0.0006797625974286348,
|
| 2490 |
+
"task": 0.17970404177904128,
|
| 2491 |
+
"count": 10
|
| 2492 |
+
}
|
| 2493 |
+
},
|
| 2494 |
+
"distill_weight": 0.26580000000000004
|
| 2495 |
+
}
|
| 2496 |
+
]
|
training_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:da40e0dea3f19b04228b37c4be914930e4240d58d39997397aaf31e334bb837e
|
| 3 |
+
size 2583
|