azliza794/myllama

README.md

@@ -18,8 +18,8 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [distilbert-base-uncased](https://huggingface.co/distilbert-base-uncased) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.8260
-- Matthews Correlation: 0.5367
+- Loss: 0.8959
+- Matthews Correlation: 0.5539
 
 ## Model description
 
@@ -38,23 +38,22 @@ More information needed
 ### Training hyperparameters
 
 The following hyperparameters were used during training:
-- learning_rate: 2e-05
-- train_batch_size: 16
+- learning_rate: 2.524386238943955e-05
+- train_batch_size: 8
 - eval_batch_size: 16
-- seed: 42
+- seed: 11
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
-- num_epochs: 5
+- num_epochs: 4
 
 ### Training results
 
 | Training Loss | Epoch | Step | Validation Loss | Matthews Correlation |
 |:-------------:|:-----:|:----:|:---------------:|:--------------------:|
-| 0.5186        | 1.0   | 535  | 0.4527          | 0.4532               |
-| 0.3437        | 2.0   | 1070 | 0.4640          | 0.5025               |
-| 0.2289        | 3.0   | 1605 | 0.6606          | 0.5127               |
-| 0.1721        | 4.0   | 2140 | 0.8260          | 0.5367               |
-| 0.1289        | 5.0   | 2675 | 0.8432          | 0.5309               |
+| 0.4835        | 1.0   | 1069 | 0.4687          | 0.4579               |
+| 0.3549        | 2.0   | 2138 | 0.5878          | 0.5113               |
+| 0.2356        | 3.0   | 3207 | 0.7891          | 0.5378               |
+| 0.1421        | 4.0   | 4276 | 0.8959          | 0.5539               |
 
 
 ### Framework versions

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:37b4ebc6ca65f593e216669994f47a9363ce75bc9f71073bffb3783be94bb86c
+oid sha256:cf56389d284dc967bd38dcf2944cabe6a3da7c03bca5a6af0caff09f7f6080d3
 size 267832560