End of training

README.md

@@ -32,7 +32,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [t5-base](https://huggingface.co/t5-base) on the glue dataset.
 It achieves the following results on the evaluation set:
-- Loss: 1.2127
+- Loss: 0.8364
 - Accuracy: 0.0
 
 ## Model description
@@ -56,8 +56,6 @@ The following hyperparameters were used during training:
 - train_batch_size: 32
 - eval_batch_size: 64
 - seed: 1
-- gradient_accumulation_steps: 2
-- total_train_batch_size: 64
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
 - lr_scheduler_warmup_steps: 20
@@ -67,14 +65,14 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch | Step | Validation Loss | Accuracy |
 |:-------------:|:-----:|:----:|:---------------:|:--------:|
-| 0.6957        | 0.64  | 25   | 0.6695          | 0.6318   |
-| 0.6356        | 1.28  | 50   | 0.6159          | 0.6751   |
-| 0.6038        | 1.92  | 75   | 0.6015          | 0.6823   |
-| 0.5251        | 2.56  | 100  | 0.5468          | 0.7581   |
-| 0.4386        | 3.21  | 125  | 0.5273          | 0.7545   |
-| 0.4175        | 3.85  | 150  | 0.5601          | 0.7545   |
-| 0.3909        | 4.49  | 175  | 0.5589          | 0.7762   |
-| 0.3342        | 5.13  | 200  | 0.5757          | 0.7581   |
+| 0.6981        | 0.32  | 25   | 0.6973          | 0.4657   |
+| 0.6818        | 0.64  | 50   | 0.6864          | 0.4910   |
+| 0.6578        | 0.96  | 75   | 0.6036          | 0.6968   |
+| 0.5986        | 1.28  | 100  | 0.5675          | 0.7292   |
+| 0.5756        | 1.6   | 125  | 0.5968          | 0.7076   |
+| 0.4797        | 1.92  | 150  | 0.5562          | 0.7401   |
+| 0.463         | 2.24  | 175  | 0.6183          | 0.7220   |
+| 0.4034        | 2.56  | 200  | 0.5857          | 0.7726   |
 
 
 ### Framework versions