jslai//content/sample_data/best_models//MBERT_uncased_CrossEntropyLoss_lora

README.md

@@ -21,12 +21,12 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [google-bert/bert-base-multilingual-uncased](https://huggingface.co/google-bert/bert-base-multilingual-uncased) on the None dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.6924
-- Accuracy: 0.518
-- F1: 0.6730
-- Precision: 0.6613
-- Recall: 0.6851
-- Roc Auc: 0.3824
+- Loss: 0.6573
+- Accuracy: 0.717
+- F1: 0.8344
+- Precision: 0.7239
+- Recall: 0.9848
+- Roc Auc: 0.4996
 
 ## Model description
 
@@ -61,9 +61,9 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch | Step | Validation Loss | Accuracy | F1     | Precision | Recall | Roc Auc |
 |:-------------:|:-----:|:----:|:---------------:|:--------:|:------:|:---------:|:------:|:-------:|
-| No log        | 0.992 | 31   | 0.7081          | 0.347    | 0.4287 | 0.5847    | 0.3384 | 0.3540  |
-| No log        | 1.984 | 62   | 0.6961          | 0.479    | 0.6354 | 0.6440    | 0.6271 | 0.3588  |
-| No log        | 2.976 | 93   | 0.6924          | 0.518    | 0.6730 | 0.6613    | 0.6851 | 0.3824  |
+| No log        | 0.992 | 31   | 0.6666          | 0.693    | 0.8174 | 0.7179    | 0.9489 | 0.4853  |
+| No log        | 1.984 | 62   | 0.6595          | 0.714    | 0.8324 | 0.7230    | 0.9807 | 0.4976  |
+| No log        | 2.976 | 93   | 0.6573          | 0.717    | 0.8344 | 0.7239    | 0.9848 | 0.4996  |
 
 
 ### Framework versions

adapter_config.json

@@ -26,22 +26,22 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "2.output.dense",
+    "9.output.dense",
+    "3.output.dense",
+    "8.output.dense",
     "10.output.dense",
     "7.output.dense",
-    "0.output.dense",
-    "11.output.dense",
-    "6.output.dense",
+    "2.output.dense",
     "query",
-    "intermediate.dense",
+    "value",
     "1.output.dense",
-    "3.output.dense",
-    "8.output.dense",
-    "key",
     "4.output.dense",
-    "value",
-    "9.output.dense",
-    "5.output.dense"
+    "intermediate.dense",
+    "5.output.dense",
+    "6.output.dense",
+    "key",
+    "0.output.dense",
+    "11.output.dense"
   ],
   "task_type": "SEQ_CLS",
   "use_dora": false,

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bbf2f8bd5fd0b9ee7d80ff96a52e17294e9a5993f0c4ff96bc133dee7c672233
+oid sha256:d34f1e1d19d119762cc1845a872ee9a1287b06575fd7f7320b3c35639db33ad3
 size 9460216