End of training

README.md

@@ -18,10 +18,11 @@ should probably proofread and complete it, then remove this comment. -->
 This model is a fine-tuned version of [google/flan-t5-small](https://huggingface.co/google/flan-t5-small) on an unknown dataset.
 It achieves the following results on the evaluation set:
 - Loss: 1.5103
-- Rouge1: 0.4692
-- Rouge2: 0.2472
-- Rougel: 0.4300
-- Rougelsum: 0.4314
+- Rouge1: 47.1409
+- Rouge2: 24.7704
+- Rougel: 43.2012
+- Rougelsum: 43.2249
+- Gen Len: 13.4509
 
 ## Model description
 
@@ -50,16 +51,16 @@ The following hyperparameters were used during training:
 
 ### Training results
 
-| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum |
-|:-------------:|:-----:|:----:|:---------------:|:------:|:------:|:------:|:---------:|
-| 1.9748        | 1.0   | 561  | 1.6071          | 0.4531 | 0.2304 | 0.4114 | 0.4114    |
-| 1.7823        | 2.0   | 1122 | 1.5643          | 0.4561 | 0.2361 | 0.4197 | 0.4202    |
-| 1.692         | 3.0   | 1683 | 1.5422          | 0.4582 | 0.2342 | 0.4210 | 0.4212    |
-| 1.6226        | 4.0   | 2244 | 1.5243          | 0.4655 | 0.2447 | 0.4288 | 0.4301    |
-| 1.5668        | 5.0   | 2805 | 1.5146          | 0.4625 | 0.2402 | 0.4257 | 0.4261    |
-| 1.5281        | 6.0   | 3366 | 1.5083          | 0.4651 | 0.2423 | 0.4293 | 0.4304    |
-| 1.5058        | 7.0   | 3927 | 1.5100          | 0.4670 | 0.2456 | 0.4290 | 0.4302    |
-| 1.4834        | 8.0   | 4488 | 1.5103          | 0.4692 | 0.2472 | 0.4300 | 0.4314    |
+| Training Loss | Epoch | Step | Validation Loss | Rouge1  | Rouge2  | Rougel  | Rougelsum | Gen Len |
+|:-------------:|:-----:|:----:|:---------------:|:-------:|:-------:|:-------:|:---------:|:-------:|
+| 1.9748        | 1.0   | 561  | 1.6071          | 45.4173 | 23.1498 | 41.2241 | 41.2184   | 14.0501 |
+| 1.7823        | 2.0   | 1122 | 1.5643          | 45.6835 | 23.6188 | 42.0517 | 42.0965   | 13.3988 |
+| 1.692         | 3.0   | 1683 | 1.5422          | 45.9189 | 23.4450 | 42.2140 | 42.2774   | 13.3287 |
+| 1.6226        | 4.0   | 2244 | 1.5243          | 46.7272 | 24.5003 | 43.0697 | 43.1200   | 13.1984 |
+| 1.5668        | 5.0   | 2805 | 1.5146          | 46.4580 | 24.0323 | 42.6975 | 42.7558   | 13.5050 |
+| 1.5281        | 6.0   | 3366 | 1.5083          | 46.7795 | 24.2760 | 43.1536 | 43.1569   | 13.5010 |
+| 1.5058        | 7.0   | 3927 | 1.5100          | 46.8698 | 24.6214 | 43.0844 | 43.0983   | 13.4509 |
+| 1.4834        | 8.0   | 4488 | 1.5103          | 47.1409 | 24.7704 | 43.2012 | 43.2249   | 13.4509 |
 
 
 ### Framework versions

generation_config.json

@@ -1,6 +1,7 @@
 {
   "decoder_start_token_id": 0,
   "eos_token_id": 1,
+  "max_new_tokens": 64,
   "pad_token_id": 0,
   "transformers_version": "4.42.4"
 }