Finished training.

README.md

@@ -23,7 +23,7 @@ model-index:
       args: irony
     metrics:
     - type: accuracy
-      value: 0.6701570680628273
+      value: 0.6743455497382199
       name: accuracy
 ---
 
@@ -34,7 +34,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [aXhyra/demo_sentiment_31415](https://huggingface.co/aXhyra/demo_sentiment_31415) on the tweet_eval dataset.
 It achieves the following results on the evaluation set:
-- accuracy: 0.6702
+- accuracy: 0.6743
 
 ## Model description
 
@@ -65,15 +65,15 @@ The following hyperparameters were used during training:
 
 | accuracy | train_loss | epoch |
 |:--------:|:----------:|:-----:|
-| 0.5204   | None       | 0     |
-| 0.6492   | 0.6611     | 0     |
-| 0.6429   | 0.6162     | 1     |
-| 0.6513   | 0.5737     | 2     |
-| 0.6754   | 0.5400     | 3     |
-| 0.6712   | 0.5228     | 4     |
-| 0.6545   | 0.5029     | 5     |
-| 0.6681   | 0.4876     | 6     |
-| 0.6702   | 0.4776     | 7     |
+| 0.4984   | None       | 0     |
+| 0.6586   | 0.6598     | 0     |
+| 0.6471   | 0.6222     | 1     |
+| 0.6597   | 0.5723     | 2     |
+| 0.6785   | 0.5469     | 3     |
+| 0.6691   | 0.5237     | 4     |
+| 0.6513   | 0.5072     | 5     |
+| 0.6702   | 0.4883     | 6     |
+| 0.6743   | 0.4823     | 7     |
 
 
 ### Framework versions

adapter_config.json

@@ -19,8 +19,8 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "q_lin",
-    "v_lin"
+    "v_lin",
+    "q_lin"
   ],
   "task_type": "SEQ_CLS",
   "use_rslora": false

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8b4feb9a5f85de17cf9fcdfa585ad35484381775b7b70c841ba1b6547eda8550
+oid sha256:71f76b445884b95d8000ecae48d724103c6c04f672b84022f59426b572e5a76d
 size 2446056