End of training

README.md

@@ -15,12 +15,12 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model was trained from scratch on the None dataset.
 It achieves the following results on the evaluation set:
-- Loss: 1.0099
-- Rouge1: 35.844
-- Rouge2: 28.0163
-- Rougel: 34.3568
-- Rougelsum: 35.3141
-- Gen Len: 25.876
+- Loss: 1.0182
+- Rouge1: 31.4073
+- Rouge2: 21.1126
+- Rougel: 27.9695
+- Rougelsum: 30.6983
+- Gen Len: 89.5188
 
 ## Model description
 
@@ -49,9 +49,9 @@ The following hyperparameters were used during training:
 
 ### Training results
 
-| Training Loss | Epoch | Step  | Validation Loss | Rouge1 | Rouge2  | Rougel  | Rougelsum | Gen Len |
-|:-------------:|:-----:|:-----:|:---------------:|:------:|:-------:|:-------:|:---------:|:-------:|
-| 2.1053        | 1.0   | 18247 | 1.0099          | 35.844 | 28.0163 | 34.3568 | 35.3141   | 25.876  |
+| Training Loss | Epoch | Step  | Validation Loss | Rouge1  | Rouge2  | Rougel  | Rougelsum | Gen Len |
+|:-------------:|:-----:|:-----:|:---------------:|:-------:|:-------:|:-------:|:---------:|:-------:|
+| 2.1392        | 1.0   | 18247 | 1.0182          | 31.4073 | 21.1126 | 27.9695 | 30.6983   | 89.5188 |
 
 
 ### Framework versions

generation_config.json

@@ -5,7 +5,7 @@
   "eos_token_id": 2,
   "forced_bos_token_id": 0,
   "forced_eos_token_id": 2,
-  "max_length": 26,
+  "max_length": 512,
   "no_repeat_ngram_size": 3,
   "num_beams": 4,
   "pad_token_id": 1,

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f58af2e64c99878b9cdf16dc006b70672118b88f67a9453e2fe17a9c6d66ffff
+oid sha256:89c7bc82292de5ae6d8f17f822c9ed730b5d257f7b095c6af40984d5b22cafb6
 size 558018637