End of training

README.md

@@ -14,7 +14,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [ybelkada/flan-t5-xl-sharded-bf16](https://huggingface.co/ybelkada/flan-t5-xl-sharded-bf16) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.2061
+- Loss: 0.3362
 
 ## Model description
 
@@ -33,24 +33,27 @@ More information needed
 ### Training hyperparameters
 
 The following hyperparameters were used during training:
-- learning_rate: 5e-05
+- learning_rate: 1e-05
 - train_batch_size: 16
 - eval_batch_size: 16
 - seed: 42
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
 - lr_scheduler_warmup_ratio: 0.1
-- num_epochs: 5
+- num_epochs: 8
 
 ### Training results
 
 | Training Loss | Epoch | Step | Validation Loss |
 |:-------------:|:-----:|:----:|:---------------:|
-| 2.6993        | 0.9   | 200  | 0.3786          |
-| 0.4329        | 1.8   | 400  | 0.2510          |
-| 0.3233        | 2.7   | 600  | 0.2303          |
-| 0.3008        | 3.6   | 800  | 0.2150          |
-| 0.2763        | 4.5   | 1000 | 0.2061          |
+| 4.6083        | 0.9   | 200  | 3.5721          |
+| 2.5382        | 1.8   | 400  | 1.3733          |
+| 1.133         | 2.7   | 600  | 0.7029          |
+| 0.7652        | 3.6   | 800  | 0.4796          |
+| 0.5812        | 4.5   | 1000 | 0.3836          |
+| 0.5035        | 5.41  | 1200 | 0.3557          |
+| 0.4842        | 6.31  | 1400 | 0.3437          |
+| 0.4591        | 7.21  | 1600 | 0.3362          |
 
 
 ### Framework versions

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0c02ebb16c738937984b2d36123a9f3513c01d9ed1d8385baa6853a22afc125f
+oid sha256:959eb5deccee0d19025879eb5d9f54246b5fae66125ad48e361439d45447f642
 size 37855242

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2767bae1f857b8913ebe0924795203fb366a8ab25cb9c0ad7b6269c507b4c520
+oid sha256:349a35d8a66dd4de58696eb8253598e256a350c3be46a96688029f6bf36a69a5
 size 5112