SvetlanaKayfajian/ss-toxicator-ru

README.md

@@ -17,12 +17,12 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [google/mt5-small](https://huggingface.co/google/mt5-small) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.0037
-- Rouge1: 7.7112
-- Rouge2: 2.5231
-- Rougel: 7.6595
-- Rougelsum: 7.7534
-- Gen Len: 16.1295
+- Loss: 0.0040
+- Rouge1: 7.6519
+- Rouge2: 2.5188
+- Rougel: 7.6693
+- Rougelsum: 7.6998
+- Gen Len: 16.1324
 
 ## Model description
 
@@ -47,22 +47,17 @@ The following hyperparameters were used during training:
 - seed: 42
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
-- num_epochs: 10
+- num_epochs: 5
 
 ### Training results
 
 | Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len |
 |:-------------:|:-----:|:----:|:---------------:|:------:|:------:|:------:|:---------:|:-------:|
-| 0.6557        | 1.0   | 348  | 0.2170          | 7.1577 | 2.1257 | 7.1089 | 7.1851    | 15.8612 |
-| 0.1692        | 2.0   | 696  | 0.0076          | 7.5739 | 2.4426 | 7.5136 | 7.5825    | 16.1374 |
-| 0.1127        | 3.0   | 1044 | 0.0076          | 7.7112 | 2.5231 | 7.6595 | 7.7534    | 16.1345 |
-| 0.073         | 4.0   | 1392 | 0.0041          | 7.7112 | 2.5231 | 7.6595 | 7.7534    | 16.1288 |
-| 0.0617        | 5.0   | 1740 | 0.0038          | 7.7112 | 2.5231 | 7.6595 | 7.7534    | 16.1309 |
-| 0.0432        | 6.0   | 2088 | 0.0034          | 7.7112 | 2.5231 | 7.6595 | 7.7534    | 16.1295 |
-| 0.0545        | 7.0   | 2436 | 0.0036          | 7.6353 | 2.4426 | 7.5819 | 7.6533    | 16.1302 |
-| 0.0349        | 8.0   | 2784 | 0.0043          | 7.7112 | 2.5231 | 7.6595 | 7.7534    | 16.1317 |
-| 0.0262        | 9.0   | 3132 | 0.0036          | 7.7112 | 2.5231 | 7.6595 | 7.7534    | 16.1302 |
-| 0.0284        | 10.0  | 3480 | 0.0037          | 7.7112 | 2.5231 | 7.6595 | 7.7534    | 16.1295 |
+| 0.6699        | 1.0   | 348  | 0.2325          | 7.206  | 2.2593 | 7.2116 | 7.2778    | 15.8029 |
+| 0.198         | 2.0   | 696  | 0.0066          | 7.5046 | 2.4469 | 7.4966 | 7.5618    | 16.1353 |
+| 0.1554        | 3.0   | 1044 | 0.0051          | 7.6519 | 2.5188 | 7.6693 | 7.6998    | 16.1309 |
+| 0.1058        | 4.0   | 1392 | 0.0039          | 7.6519 | 2.5188 | 7.6693 | 7.6998    | 16.1317 |
+| 0.1013        | 5.0   | 1740 | 0.0040          | 7.6519 | 2.5188 | 7.6693 | 7.6998    | 16.1324 |
 
 
 ### Framework versions

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5b40e9ad70ee7d36b7c37b20045cdefcba31535bb88aea3406b417217442ae53
+oid sha256:9be5265e4c6305960c6970e2c32f7a92e8fdf98f2ed8c0b6d5dfee91c4cf646b
 size 1200729512

runs/May12_09-02-34_8b6b0d2a77b6/events.out.tfevents.1715504573.8b6b0d2a77b6.1287.0

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:11df5281ce3e274e49c85c6f89f94273c4cd456d28806e7718f4f58c8cf20f46
-size 70948
+oid sha256:8522b9fe6c9a6a139c23472a536a469fc5367bd85eab7a5aee498d32beb05948
+size 81322

tokenizer.json

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fc78c17ee5533a6bf363bdfb004077863736668af937403cd3e8905b32a4630e
-size 16330638
+oid sha256:ddb1b00035edffde435e24e69011b7f552691405237455cb7c3a59c890682807
+size 16330540