End of training

README.md

@@ -15,7 +15,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [distilgpt2](https://huggingface.co/distilgpt2) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 2.6367
+- Loss: 2.2461
 
 ## Model description
 
@@ -35,78 +35,27 @@ More information needed
 
 The following hyperparameters were used during training:
 - learning_rate: 0.0005
-- train_batch_size: 24
-- eval_batch_size: 24
+- train_batch_size: 4
+- eval_batch_size: 4
 - seed: 42
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: cosine
 - lr_scheduler_warmup_steps: 1000
-- num_epochs: 40
+- num_epochs: 1
 
 ### Training results
 
-| Training Loss | Epoch | Step | Validation Loss |
-|:-------------:|:-----:|:-----:|:---------------:|
-| 3.6682 | 0.66 | 1000 | 2.7578 |
-| 3.0752 | 1.32 | 2000 | 2.6973 |
-| 3.0293 | 1.99 | 3000 | 2.7012 |
-| 3.0137 | 2.65 | 4000 | 2.6465 |
-| 3.0057 | 3.31 | 5000 | 2.6562 |
-| 3.0113 | 3.97 | 6000 | 2.6621 |
-| 3.0044 | 4.64 | 7000 | 2.6426 |
-| 3.0059 | 5.3 | 8000 | 2.6719 |
-| 3.0087 | 5.96 | 9000 | 2.6602 |
-| 3.0002 | 6.62 | 10000 | 2.6406 |
-| 2.9971 | 7.28 | 11000 | 2.6680 |
-| 2.9896 | 7.95 | 12000 | 2.6602 |
-| 2.9936 | 8.61 | 13000 | 2.6699 |
-| 2.9985 | 9.27 | 14000 | 2.6641 |
-| 2.9845 | 9.93 | 15000 | 2.6660 |
-| 2.9953 | 10.6 | 16000 | 2.6523 |
-| 2.9858 | 11.26 | 17000 | 2.6621 |
-| 2.9892 | 11.92 | 18000 | 2.6445 |
-| 2.9869 | 12.58 | 19000 | 2.625 |
-| 2.9899 | 13.25 | 20000 | 2.6465 |
-| 2.9824 | 13.91 | 21000 | 2.6836 |
-| 2.9824 | 14.57 | 22000 | 2.6445 |
-| 2.9869 | 15.23 | 23000 | 2.6641 |
-| 2.9864 | 15.89 | 24000 | 2.6543 |
-| 2.9815 | 16.56 | 25000 | 2.6309 |
-| 2.9806 | 17.22 | 26000 | 2.6777 |
-| 2.9884 | 17.88 | 27000 | 2.6270 |
-| 2.9919 | 18.54 | 28000 | 2.6445 |
-| 2.9842 | 19.21 | 29000 | 2.6758 |
-| 2.9771 | 19.87 | 30000 | 2.6543 |
-| 2.9824 | 20.53 | 31000 | 2.6523 |
-| 2.9889 | 21.19 | 32000 | 2.6523 |
-| 2.9816 | 21.85 | 33000 | 2.6270 |
-| 2.9844 | 22.52 | 34000 | 2.6562 |
-| 2.9781 | 23.18 | 35000 | 2.6328 |
-| 2.9802 | 23.84 | 36000 | 2.6289 |
-| 2.9783 | 24.5 | 37000 | 2.6270 |
-| 2.9832 | 25.17 | 38000 | 2.6406 |
-| 2.9759 | 25.83 | 39000 | 2.6289 |
-| 2.9805 | 26.49 | 40000 | 2.6387 |
-| 2.9795 | 27.15 | 41000 | 2.6367 |
-| 2.9728 | 27.81 | 42000 | 2.6367 |
-| 2.978 | 28.48 | 43000 | 2.6426 |
-| 2.9779 | 29.14 | 44000 | 2.6348 |
-| 2.977 | 29.8 | 45000 | 2.6406 |
-| 2.9692 | 30.46 | 46000 | 2.6348 |
-| 2.9815 | 31.13 | 47000 | 2.6406 |
-| 2.9706 | 31.79 | 48000 | 2.6387 |
-| 2.9764 | 32.45 | 49000 | 2.6367 |
-| 2.9733 | 33.11 | 50000 | 2.6230 |
-| 2.9751 | 33.77 | 51000 | 2.6504 |
-| 2.968 | 34.44 | 52000 | 2.6465 |
-| 2.972 | 35.1 | 53000 | 2.6523 |
-| 2.9744 | 35.76 | 54000 | 2.6406 |
-| 2.9611 | 36.42 | 55000 | 2.6328 |
-| 2.9786 | 37.09 | 56000 | 2.6348 |
-| 2.9712 | 37.75 | 57000 | 2.6406 |
-| 2.9722 | 38.41 | 58000 | 2.6348 |
-| 2.9709 | 39.07 | 59000 | 2.6367 |
-| 2.9735 | 39.74 | 60000 | 2.6367 |
+| Training Loss | Epoch | Step | Validation Loss |
+|:-------------:|:-----:|:----:|:---------------:|
+| 2.9811 | 0.11 | 1000 | 2.4141 |
+| 2.5312 | 0.22 | 2000 | 2.3164 |
+| 2.4908 | 0.33 | 3000 | 2.2871 |
+| 2.4785 | 0.44 | 4000 | 2.2754 |
+| 2.4518 | 0.55 | 5000 | 2.2832 |
+| 2.4277 | 0.66 | 6000 | 2.2578 |
+| 2.4352 | 0.77 | 7000 | 2.25 |
+| 2.4171 | 0.88 | 8000 | 2.2480 |
+| 2.4138 | 0.99 | 9000 | 2.2461 |
 
 
 ### Framework versions

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e22809645a4aafe8f25ef37433cb46d18e7bed28a4b2ca946e7d93ed00b1e49e
+oid sha256:0bd87c89fcd8fe6d7f4864e45015714123e4b18094325f9796b81192c55eec2a
 size 418013

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d92186ffc4adc2ba0962f503328a317336dcea34dd6748736f99e51aaac2c7cb
+oid sha256:e80b6fef053a79b4d6a8998a5ba2c4f54d311285bd63be90fb5dd6dc44493072
 size 4091