End of training

README.md

@@ -18,10 +18,10 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [meta-llama/Llama-3.2-3B](https://huggingface.co/meta-llama/Llama-3.2-3B) on the None dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.4937
+- Loss: 0.3049
 - Model Preparation Time: 0.0047
-- Accuracy: 0.7875
-- F1 Macro: 0.7871
+- Accuracy: 0.9102
+- F1 Macro: 0.9104
 
 ## Model description
 
@@ -56,12 +56,10 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch | Step | Validation Loss | Model Preparation Time | Accuracy | F1 Macro |
 |:-------------:|:-----:|:----:|:---------------:|:----------------------:|:--------:|:--------:|
-| 1.2728        | 1.0   | 160  | 1.2319          | 0.0047                 | 0.4672   | 0.4688   |
-| 0.7863        | 2.0   | 320  | 0.6863          | 0.0047                 | 0.7297   | 0.7298   |
-| 0.5238        | 3.0   | 480  | 0.5408          | 0.0047                 | 0.7703   | 0.7693   |
-| 0.4253        | 4.0   | 640  | 0.5223          | 0.0047                 | 0.7906   | 0.7906   |
-| 0.4088        | 5.0   | 800  | 0.5165          | 0.0047                 | 0.7969   | 0.7966   |
-| 0.3496        | 6.0   | 960  | 0.5318          | 0.0047                 | 0.7891   | 0.7882   |
+| 0.8579        | 1.0   | 270  | 0.7867          | 0.0047                 | 0.7167   | 0.7161   |
+| 0.2739        | 2.0   | 540  | 0.3057          | 0.0047                 | 0.9037   | 0.9036   |
+| 0.2183        | 3.0   | 810  | 0.2756          | 0.0047                 | 0.9111   | 0.9106   |
+| 0.1703        | 4.0   | 1080 | 0.2870          | 0.0047                 | 0.8972   | 0.8963   |
 
 
 ### Framework versions

adapter_config.json

@@ -26,10 +26,10 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "v_proj",
     "q_proj",
-    "k_proj",
-    "o_proj"
+    "o_proj",
+    "v_proj",
+    "k_proj"
   ],
   "task_type": "SEQ_CLS",
   "use_dora": false,

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7687cbab4b7dc6090289678c78f45a90ffcc16513ff4c47591855c3edf728ff2
+oid sha256:cc84cc5602adf9457600966fa47be62790703dfe9530f3a2df2cb6706d5c9444
 size 36779480