Finished training.

README.md

@@ -23,7 +23,7 @@ model-index:
       args: default
     metrics:
     - type: accuracy
-      value: 0.9344736842105263
+      value: 0.9335526315789474
       name: accuracy
 ---
 
@@ -34,7 +34,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [anirudh21/bert-base-uncased-finetuned-qnli](https://huggingface.co/anirudh21/bert-base-uncased-finetuned-qnli) on the ag_news dataset.
 It achieves the following results on the evaluation set:
-- accuracy: 0.9345
+- accuracy: 0.9336
 
 ## Model description
 
@@ -65,11 +65,11 @@ The following hyperparameters were used during training:
 
 | accuracy | train_loss | epoch |
 |:--------:|:----------:|:-----:|
-| 0.3028   | None       | 0     |
-| 0.9249   | 0.2906     | 0     |
-| 0.9267   | 0.2142     | 1     |
-| 0.9326   | 0.1963     | 2     |
-| 0.9345   | 0.1868     | 3     |
+| 0.2609   | None       | 0     |
+| 0.9232   | 0.2931     | 0     |
+| 0.9275   | 0.2171     | 1     |
+| 0.9322   | 0.1981     | 2     |
+| 0.9336   | 0.1883     | 3     |
 
 
 ### Framework versions

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c60a2baefb1ef4f25ffee820e80b2101d49529efa982ecaf28d86d83a381a48f
+oid sha256:77d0700cd5a805f58a471119df6f47e46167f458732625f90f854b541be4cb1c
 size 166552