ninagroot/GPT2-705Mtest

README.md

@@ -13,7 +13,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [](https://huggingface.co/) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 7.2081
+- Loss: 6.4596
 
 ## Model description
 
@@ -41,22 +41,25 @@ The following hyperparameters were used during training:
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: cosine
 - lr_scheduler_warmup_steps: 50
-- num_epochs: 15
+- num_epochs: 20
 - mixed_precision_training: Native AMP
 
 ### Training results
 
 | Training Loss | Epoch | Step | Validation Loss |
 |:-------------:|:-----:|:----:|:---------------:|
-| 9.6439        | 0.57  | 1    | 9.6409          |
-| 8.0775        | 1.71  | 3    | 9.6248          |
-| 8.2418        | 2.86  | 5    | 8.7251          |
-| 7.6995        | 4.0   | 7    | 8.0438          |
-| 7.4715        | 4.57  | 8    | 8.1243          |
-| 7.7918        | 5.71  | 10   | 7.7544          |
-| 7.1086        | 6.86  | 12   | 7.4231          |
-| 6.7919        | 8.0   | 14   | 7.1621          |
-| 6.4626        | 8.57  | 15   | 7.2081          |
+| 9.6842        | 0.57  | 1    | 9.6923          |
+| 7.965         | 1.71  | 3    | 8.8509          |
+| 7.7822        | 2.86  | 5    | 8.3252          |
+| 7.8086        | 4.0   | 7    | 7.8650          |
+| 7.3927        | 4.57  | 8    | 7.6373          |
+| 7.3496        | 5.71  | 10   | 7.5233          |
+| 6.7868        | 6.86  | 12   | 7.0954          |
+| 7.3518        | 8.0   | 14   | 7.1324          |
+| 6.4308        | 8.57  | 15   | 6.8810          |
+| 6.0271        | 9.71  | 17   | 6.6660          |
+| 5.7575        | 10.86 | 19   | 6.5046          |
+| 5.7538        | 11.43 | 20   | 6.4596          |
 
 
 ### Framework versions

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2f1f9203db249777c29b82b484e27c1fc2e664be0559a016f326bdee7f55265f
+oid sha256:b6da372f2e73ec6c170255625319c33708e0fa2290cc9329875f837849601ec0
 size 2796386080

runs/Apr17_10-08-09_gcn48.local.snellius.surf.nl/events.out.tfevents.1713341298.gcn48.local.snellius.surf.nl.409298.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1a5f511acee527a6412b8b58c5b8ce5fff939cd6337934acb235ed6d16c60302
+size 12358

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6e54f97170740d4086e8930b99a3fa778220e4cb0c9979b0c991f4a96c4a2129
+oid sha256:b7460a82fa360eea14e5e28ddaf263930380d8739c0dcc5e38456889b2b2ef9f
 size 4984