Colabng/twitter_bank_scam_classifier

README.md

@@ -20,9 +20,9 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [google-bert/bert-base-uncased](https://huggingface.co/google-bert/bert-base-uncased) on the None dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.7566
-- Accuracy: 0.64
-- Auc: 0.67
+- Loss: 0.7791
+- Accuracy: 0.5
+- Auc: 0.55
 
 ## Model description
 
@@ -53,14 +53,14 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch | Step | Validation Loss | Accuracy | Auc  |
 |:-------------:|:-----:|:----:|:---------------:|:--------:|:----:|
-| 0.7449        | 1.0   | 11   | 0.7669          | 0.5      | 0.66 |
-| 0.7084        | 2.0   | 22   | 0.7566          | 0.64     | 0.67 |
+| 0.7848        | 1.0   | 11   | 0.7885          | 0.45     | 0.51 |
+| 0.633         | 2.0   | 22   | 0.7791          | 0.5      | 0.55 |
 
 
 ### Framework versions
 
 - PEFT 0.16.0
 - Transformers 4.53.3
-- Pytorch 2.7.1+cpu
+- Pytorch 2.7.1+cu126
 - Datasets 4.0.0
 - Tokenizers 0.21.2

adapter_config.json

@@ -28,8 +28,8 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "key",
     "query",
+    "key",
     "value"
   ],
   "task_type": "SEQ_CLS",

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fd6ab6fde9b36810e90f70609a7a3ee618b9b69b51b0a9408922da72f0b9a579
+oid sha256:5580fd44bf139499264952161a4a99221479e20c6483be68ca4f2f93f18d9307
 size 95583552

runs/Sep07_02-05-34_job-d2uefgre5dus73em5ce0-lkc94/events.out.tfevents.1757210735.job-d2uefgre5dus73em5ce0-lkc94

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ed5c45166c32e2920438a6d47fb5f72e4a085c32c6890c70c29745167e295839
+size 6555

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4daa0d8d24dab83655ff5277850650b8124d530a5ed73a0f7f4b2731a444325c
+oid sha256:50642fb26aa03cdce1003e6488c68fdfe7140baf3fc4d0e9dd5b52473ebf4262
 size 5777