Model save

README.md

@@ -16,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [gpt2](https://huggingface.co/gpt2) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.7416
+- Loss: 0.5208
 
 ## Model description
 
@@ -41,16 +41,51 @@ The following hyperparameters were used during training:
 - seed: 42
 - optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
 - lr_scheduler_type: linear
-- num_epochs: 3
+- num_epochs: 50
 - mixed_precision_training: Native AMP
 
 ### Training results
 
 | Training Loss | Epoch | Step | Validation Loss |
 |:-------------:|:-----:|:----:|:---------------:|
-| 0.4261        | 1.0   | 237  | 1.0854          |
-| 0.2515        | 2.0   | 474  | 0.7846          |
-| 0.2148        | 3.0   | 711  | 0.7416          |
+| 0.3959        | 1.0   | 237  | 0.9383          |
+| 0.2182        | 2.0   | 474  | 0.6835          |
+| 0.1772        | 3.0   | 711  | 0.6222          |
+| 0.1676        | 4.0   | 948  | 0.5980          |
+| 0.1609        | 5.0   | 1185 | 0.5840          |
+| 0.1519        | 6.0   | 1422 | 0.5740          |
+| 0.1535        | 7.0   | 1659 | 0.5659          |
+| 0.1492        | 8.0   | 1896 | 0.5587          |
+| 0.1456        | 9.0   | 2133 | 0.5575          |
+| 0.1427        | 10.0  | 2370 | 0.5528          |
+| 0.1442        | 11.0  | 2607 | 0.5499          |
+| 0.1403        | 12.0  | 2844 | 0.5466          |
+| 0.1425        | 13.0  | 3081 | 0.5449          |
+| 0.1403        | 14.0  | 3318 | 0.5414          |
+| 0.1405        | 15.0  | 3555 | 0.5399          |
+| 0.1387        | 16.0  | 3792 | 0.5383          |
+| 0.1377        | 17.0  | 4029 | 0.5368          |
+| 0.1379        | 18.0  | 4266 | 0.5376          |
+| 0.1353        | 19.0  | 4503 | 0.5349          |
+| 0.1378        | 20.0  | 4740 | 0.5313          |
+| 0.1372        | 21.0  | 4977 | 0.5320          |
+| 0.135         | 22.0  | 5214 | 0.5286          |
+| 0.1361        | 23.0  | 5451 | 0.5282          |
+| 0.1357        | 24.0  | 5688 | 0.5287          |
+| 0.1372        | 25.0  | 5925 | 0.5269          |
+| 0.1343        | 26.0  | 6162 | 0.5271          |
+| 0.1321        | 27.0  | 6399 | 0.5255          |
+| 0.1341        | 28.0  | 6636 | 0.5240          |
+| 0.1346        | 29.0  | 6873 | 0.5235          |
+| 0.1327        | 30.0  | 7110 | 0.5239          |
+| 0.1335        | 31.0  | 7347 | 0.5230          |
+| 0.1332        | 32.0  | 7584 | 0.5228          |
+| 0.1317        | 33.0  | 7821 | 0.5234          |
+| 0.1331        | 34.0  | 8058 | 0.5220          |
+| 0.1327        | 35.0  | 8295 | 0.5213          |
+| 0.1338        | 36.0  | 8532 | 0.5204          |
+| 0.1306        | 37.0  | 8769 | 0.5206          |
+| 0.1317        | 38.0  | 9006 | 0.5208          |
 
 
 ### Framework versions

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:16c8931d0bb7a2c11735927e9ac98e1995f962bb42247bf70b5e2502799143fa
+oid sha256:7e6e14a89bed03dbce4dac616e7c984788a01722d603d8d2a3c8da33137a8f91
 size 309974336

all_results.json

@@ -1,13 +1,13 @@
 {
-    "epoch": 3.0,
+    "epoch": 38.0,
     "eval_loss": 0.7415885329246521,
     "eval_runtime": 9.8328,
     "eval_samples_per_second": 161.399,
     "eval_steps_per_second": 5.085,
     "perplexity": 2.0992676229483833,
-    "total_flos": 1489415748452352.0,
-    "train_loss": 0.3230170006490458,
-    "train_runtime": 81.3392,
-    "train_samples_per_second": 279.349,
-    "train_steps_per_second": 8.741
+    "total_flos": 1.886593281372979e+16,
+    "train_loss": 0.15147609674822032,
+    "train_runtime": 971.6736,
+    "train_samples_per_second": 389.74,
+    "train_steps_per_second": 12.195
 }

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 3.0,
-    "total_flos": 1489415748452352.0,
-    "train_loss": 0.3230170006490458,
-    "train_runtime": 81.3392,
-    "train_samples_per_second": 279.349,
-    "train_steps_per_second": 8.741
+    "epoch": 38.0,
+    "total_flos": 1.886593281372979e+16,
+    "train_loss": 0.15147609674822032,
+    "train_runtime": 971.6736,
+    "train_samples_per_second": 389.74,
+    "train_steps_per_second": 12.195
 }