End of training

README.md

@@ -17,9 +17,9 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [google-t5/t5-small](https://huggingface.co/google-t5/t5-small) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 1.4800
-- Bleu: 6.565
-- Gen Len: 17.5641
+- Loss: 1.5191
+- Bleu: 6.3813
+- Gen Len: 17.539
 
 ## Model description
 
@@ -38,22 +38,24 @@ More information needed
 ### Training hyperparameters
 
 The following hyperparameters were used during training:
-- learning_rate: 0.0001
-- train_batch_size: 32
-- eval_batch_size: 32
+- learning_rate: 2e-05
+- train_batch_size: 16
+- eval_batch_size: 16
 - seed: 42
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
-- num_epochs: 3
+- num_epochs: 5
 - mixed_precision_training: Native AMP
 
 ### Training results
 
-| Training Loss | Epoch | Step | Validation Loss | Bleu   | Gen Len |
-|:-------------:|:-----:|:----:|:---------------:|:------:|:-------:|
-| 1.7648        | 1.0   | 2542 | 1.5394          | 6.1939 | 17.572  |
-| 1.6761        | 2.0   | 5084 | 1.4921          | 6.5053 | 17.5665 |
-| 1.6496        | 3.0   | 7626 | 1.4800          | 6.565  | 17.5641 |
+| Training Loss | Epoch | Step  | Validation Loss | Bleu   | Gen Len |
+|:-------------:|:-----:|:-----:|:---------------:|:------:|:-------:|
+| 1.8456        | 1.0   | 6355  | 1.6112          | 5.7972 | 17.5672 |
+| 1.7857        | 2.0   | 12710 | 1.5620          | 6.1557 | 17.5515 |
+| 1.7359        | 3.0   | 19065 | 1.5358          | 6.2797 | 17.5462 |
+| 1.7219        | 4.0   | 25420 | 1.5226          | 6.3581 | 17.5427 |
+| 1.7219        | 5.0   | 31775 | 1.5191          | 6.3813 | 17.539  |
 
 
 ### Framework versions

config.json

@@ -55,7 +55,7 @@
     }
   },
   "torch_dtype": "float32",
-  "transformers_version": "4.39.0",
+  "transformers_version": "4.40.1",
   "use_cache": true,
   "vocab_size": 32128
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f078b019febfa528c2d1afc03f6a276881c487813d399020b0e687394e9468fa
+oid sha256:47a1a88f77a33b53d7f668affbc94b5de6cb8427cd76efd6bfdf358bb1f20a52
 size 242041896

runs/May01_21-22-28_a8d4deb0cc84/events.out.tfevents.1714598549.a8d4deb0cc84.153.0

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:de403ac9666a5fff081a86807211d59104d527795fbcdd95a55b5f5deffedc31
-size 20606
+oid sha256:db19c5cb7f54e6aed51283cca67bf4e09677915c239c96e6495f28ab17c6c21e
+size 21343

tokenizer.json

@@ -949,8 +949,8 @@
       {
         "type": "Metaspace",
         "replacement": "▁",
-        "add_prefix_space": true,
-        "prepend_scheme": "always"
+        "prepend_scheme": "always",
+        "split": true
       }
     ]
   },
@@ -1011,8 +1011,8 @@
   "decoder": {
     "type": "Metaspace",
     "replacement": "▁",
-    "add_prefix_space": true,
-    "prepend_scheme": "always"
+    "prepend_scheme": "always",
+    "split": true
   },
   "model": {
     "type": "Unigram",

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ff3cfa8fd6e9a29141be3badbe1e4bd488ec01de78c7db62942081f8ecceb5cf
-size 5048
+oid sha256:b45c36594ab696d7b9b4d714230a85c2bfc0cbc438654bf1398a104f054fab7c
+size 5176