Downloads · 30 days
54
1% of all-time downloads
msft-regular-model
mikaelsouza/msft-regular-model
msft-regular-model is a text generation model from mikaelsouza. Use it when you need the model to write or continue text. It is set up for transformers.
should probably proofread and complete it, then remove this comment. --
All-time downloads
10.2K
Public
Repo size
164 MB
Likes
1
Public
Repo makeup
Click a slice to open those files.
.bin82 MB · 97%
At a glance
- Task
- Text Generation
- Library
- transformers
- Model type
- gpt2
- Access
- Public
- Created
- Mar 2, 2022
- Updated
- Nov 2, 2021
- SHA
- 0bec60c9
Try a prompt
Datasets
Model card
From the Hugging Face model README
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
msft-regular-model
This model is a fine-tuned version of on the wikitext dataset. It achieves the following results on the evaluation set:
- Loss: 5.3420
Model description
More information needed
Intended uses & limitations
More information needed
Training and evaluation data
More information needed
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 5e-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: 20
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 9.1224 | 0.17 | 200 | 8.0736 |
| 7.5229 | 0.34 | 400 | 7.1536 |
| 7.0122 | 0.51 | 600 | 6.9072 |
| 6.8296 | 0.69 | 800 | 6.7582 |
| 6.709 | 0.86 | 1000 | 6.6436 |
| 6.5882 | 1.03 | 1200 | 6.5563 |
| 6.4807 | 1.2 | 1400 | 6.4784 |
| 6.4172 | 1.37 | 1600 | 6.4165 |
| 6.3403 | 1.54 | 1800 | 6.3555 |
| 6.2969 | 1.71 | 2000 | 6.3107 |
| 6.2346 | 1.89 | 2200 | 6.2691 |
| 6.1767 | 2.06 | 2400 | 6.2299 |
| 6.1326 | 2.23 | 2600 | 6.1937 |
| 6.1035 | 2.4 | 2800 | 6.1602 |
| 6.0624 | 2.57 | 3000 | 6.1241 |
| 6.0393 | 2.74 | 3200 | 6.0971 |
| 5.9982 | 2.91 | 3400 | 6.0656 |
| 5.9526 | 3.08 | 3600 | 6.0397 |
| 5.9086 | 3.26 | 3800 | 6.0104 |
| 5.8922 | 3.43 | 4000 | 5.9888 |
| 5.8631 | 3.6 | 4200 | 5.9661 |
| 5.8396 | 3.77 | 4400 | 5.9407 |
| 5.8055 | 3.94 | 4600 | 5.9177 |
| 5.7763 | 4.11 | 4800 | 5.9007 |
| 5.7314 | 4.28 | 5000 | 5.8834 |
| 5.7302 | 4.46 | 5200 | 5.8620 |
| 5.6987 | 4.63 | 5400 | 5.8451 |
| 5.6754 | 4.8 | 5600 | 5.8242 |
| 5.6571 | 4.97 | 5800 | 5.8059 |
| 5.615 | 5.14 | 6000 | 5.7871 |
| 5.596 | 5.31 | 6200 | 5.7817 |
| 5.5738 | 5.48 | 6400 | 5.7570 |
| 5.5641 | 5.66 | 6600 | 5.7431 |
| 5.5503 | 5.83 | 6800 | 5.7271 |
| 5.5214 | 6.0 | 7000 | 5.7108 |
| 5.4712 | 6.17 | 7200 | 5.7018 |
| 5.48 | 6.34 | 7400 | 5.6936 |
| 5.4527 | 6.51 | 7600 | 5.6812 |
| 5.4514 | 6.68 | 7800 | 5.6669 |
| 5.4454 | 6.86 | 8000 | 5.6509 |
| 5.399 | 7.03 | 8200 | 5.6408 |
| 5.3747 | 7.2 | 8400 | 5.6327 |
| 5.3667 | 7.37 | 8600 | 5.6197 |
| 5.3652 | 7.54 | 8800 | 5.6084 |
| 5.3394 | 7.71 | 9000 | 5.5968 |
| 5.3349 | 7.88 | 9200 | 5.5870 |
| 5.2994 | 8.05 | 9400 | 5.5826 |
| 5.2793 | 8.23 | 9600 | 5.5710 |
| 5.2716 | 8.4 | 9800 | 5.5623 |
| 5.275 | 8.57 | 10000 | 5.5492 |
| 5.264 | 8.74 | 10200 | 5.5449 |
| 5.241 | 8.91 | 10400 | 5.5322 |
| 5.2285 | 9.08 | 10600 | 5.5267 |
| 5.2021 | 9.25 | 10800 | 5.5187 |
| 5.1934 | 9.43 | 11000 | 5.5158 |
| 5.1737 | 9.6 | 11200 | 5.5044 |
| 5.1774 | 9.77 | 11400 | 5.5008 |
| 5.1841 | 9.94 | 11600 | 5.4960 |
| 5.1414 | 10.11 | 11800 | 5.4895 |
| 5.1491 | 10.28 | 12000 | 5.4849 |
| 5.1184 | 10.45 | 12200 | 5.4738 |
| 5.1136 | 10.63 | 12400 | 5.4690 |
| 5.1199 | 10.8 | 12600 | 5.4598 |
| 5.1056 | 10.97 | 12800 | 5.4536 |
| 5.0648 | 11.14 | 13000 | 5.4496 |
| 5.0598 | 11.31 | 13200 | 5.4449 |
| 5.0656 | 11.48 | 13400 | 5.4422 |
| 5.0664 | 11.65 | 13600 | 5.4367 |
| 5.0675 | 11.83 | 13800 | 5.4286 |
| 5.0459 | 12.0 | 14000 | 5.4249 |
| 5.0073 | 12.17 | 14200 | 5.4260 |
| 5.0229 | 12.34 | 14400 | 5.4175 |
| 5.0079 | 12.51 | 14600 | 5.4119 |
| 5.0 | 12.68 | 14800 | 5.4194 |
| 5.0094 | 12.85 | 15000 | 5.4068 |
| 4.9967 | 13.02 | 15200 | 5.3995 |
| 4.9541 | 13.2 | 15400 | 5.4002 |
| 4.9753 | 13.37 | 15600 | 5.3965 |
| 4.9732 | 13.54 | 15800 | 5.3925 |
| 4.9624 | 13.71 | 16000 | 5.3888 |
| 4.9559 | 13.88 | 16200 | 5.3824 |
| 4.9559 | 14.05 | 16400 | 5.3851 |
| 4.9109 | 14.22 | 16600 | 5.3815 |
| 4.9211 | 14.4 | 16800 | 5.3784 |
| 4.9342 | 14.57 | 17000 | 5.3735 |
| 4.9271 | 14.74 | 17200 | 5.3711 |
| 4.9328 | 14.91 | 17400 | 5.3646 |
| 4.8994 | 15.08 | 17600 | 5.3664 |
| 4.8932 | 15.25 | 17800 | 5.3642 |
| 4.8886 | 15.42 | 18000 | 5.3620 |
| 4.8997 | 15.6 | 18200 | 5.3584 |
| 4.8846 | 15.77 | 18400 | 5.3551 |
| 4.8993 | 15.94 | 18600 | 5.3516 |
| 4.8648 | 16.11 | 18800 | 5.3552 |
| 4.8838 | 16.28 | 19000 | 5.3512 |
| 4.8575 | 16.45 | 19200 | 5.3478 |
| 4.8623 | 16.62 | 19400 | 5.3480 |
| 4.8631 | 16.8 | 19600 | 5.3439 |
| 4.8576 | 16.97 | 19800 | 5.3428 |
| 4.8265 | 17.14 | 20000 | 5.3420 |
| 4.8523 | 17.31 | 20200 | 5.3410 |
| 4.8477 | 17.48 | 20400 | 5.3396 |
| 4.8507 | 17.65 | 20600 | 5.3380 |
| 4.8498 | 17.82 | 20800 | 5.3333 |
| 4.8261 | 17.99 | 21000 | 5.3342 |
| 4.8201 | 18.17 | 21200 | 5.3324 |
| 4.8214 | 18.34 | 21400 | 5.3341 |
| 4.8195 | 18.51 | 21600 | 5.3315 |
| 4.8216 | 18.68 | 21800 | 5.3335 |
| 4.8243 | 18.85 | 22000 | 5.3291 |
| 4.832 | 19.02 | 22200 | 5.3295 |
| 4.8085 | 19.19 | 22400 | 5.3309 |
| 4.8094 | 19.37 | 22600 | 5.3283 |
| 4.815 | 19.54 | 22800 | 5.3280 |
| 4.8219 | 19.71 | 23000 | 5.3270 |
| 4.8117 | 19.88 | 23200 | 5.3280 |
Framework versions
- Transformers 4.13.0.dev0
- Pytorch 1.10.0
- Datasets 1.14.0
- Tokenizers 0.10.3
- Task
- Text Generation
- Library
- transformers
- Type
- gpt2
- Created
- Mar 2, 2022
- Updated
- Nov 2, 2021