Update README.md
Browse files
README.md
CHANGED
@@ -13,9 +13,9 @@ should probably proofread and complete it, then remove this comment. -->
|
|
13 |
|
14 |
# Wikibot-3001
|
15 |
|
16 |
-
This model is a fine-tuned version of [
|
17 |
It achieves the following results on the evaluation set:
|
18 |
-
- Loss: 1.
|
19 |
|
20 |
## Model description
|
21 |
|
@@ -41,15 +41,24 @@ The following hyperparameters were used during training:
|
|
41 |
- distributed_type: tpu
|
42 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
43 |
- lr_scheduler_type: linear
|
44 |
-
- num_epochs:
|
45 |
|
46 |
### Training results
|
47 |
|
48 |
| Training Loss | Epoch | Step | Validation Loss |
|
49 |
|:-------------:|:-----:|:-----:|:---------------:|
|
50 |
-
|
|
51 |
-
|
|
52 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
53 |
|
54 |
|
55 |
### Framework versions
|
|
|
13 |
|
14 |
# Wikibot-3001
|
15 |
|
16 |
+
This model is a fine-tuned version of [gpt2](https://huggingface.co/gpt2) on the [mapama247/wikihow_es](https://huggingface.co/datasets/mapama247/wikihow_es) dataset.
|
17 |
It achieves the following results on the evaluation set:
|
18 |
+
- Loss: 1.9422
|
19 |
|
20 |
## Model description
|
21 |
|
|
|
41 |
- distributed_type: tpu
|
42 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
43 |
- lr_scheduler_type: linear
|
44 |
+
- num_epochs: 12
|
45 |
|
46 |
### Training results
|
47 |
|
48 |
| Training Loss | Epoch | Step | Validation Loss |
|
49 |
|:-------------:|:-----:|:-----:|:---------------:|
|
50 |
+
| 2.7432 | 1.0 | 5950 | 2.6109 |
|
51 |
+
| 2.3556 | 2.0 | 11900 | 2.3059 |
|
52 |
+
| 2.1872 | 3.0 | 17850 | 2.1738 |
|
53 |
+
| 2.0891 | 4.0 | 23800 | 2.0990 |
|
54 |
+
| 2.0306 | 5.0 | 29750 | 2.0579 |
|
55 |
+
| 1.9975 | 6.0 | 35700 | 2.0395 |
|
56 |
+
| 1.9909 | 7.0 | 5950 | 2.0208 |
|
57 |
+
| 1.8924 | 8.0 | 11900 | 1.9678 |
|
58 |
+
| 1.8281 | 9.0 | 17850 | 1.9422 |
|
59 |
+
| 1.8376 | 10.0 | 5950 | 1.9576 |
|
60 |
+
| 1.7802 | 11.0 | 11900 | 1.9212 |
|
61 |
+
| 1.7465 | 12.0 | 17850 | 1.8996 |
|
62 |
|
63 |
|
64 |
### Framework versions
|