Downloads · 30 days
2
5% of all-time downloads
Hyponatremia_M2_1000steps_1e5rate_03beta_DPO
tsavage68/Hyponatremia_M2_1000steps_1e5rate_03beta_DPO
Hyponatremia_M2_1000steps_1e5rate_03beta_DPO is a text generation model from tsavage68. Use it when you need the model to write or continue text. It is set up for transformers. The card lists the license as apache-2.0.
should probably proofread and complete it, then remove this comment. --
All-time downloads
41
Public
Parameters
7.2B
14.5 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors29 GB · 100%
At a glance
- Task
- Text Generation
- Library
- transformers
- License
- apache-2.0
- Model type
- mistral
- Access
- Public
- Created
- Jul 12, 2024
- Updated
- Jul 12, 2024
- SHA
- 5475f97f
Try a prompt
Base models
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. -->
Hyponatremia_M2_1000steps_1e5rate_03beta_DPO
This model is a fine-tuned version of tsavage68/Hyponatremia_M2_150steps_1e6rate_SFT on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.0000
- Rewards/chosen: 1.3061
- Rewards/rejected: -19.7637
- Rewards/accuracies: 1.0
- Rewards/margins: 21.0698
- Logps/rejected: -136.7606
- Logps/chosen: -32.3735
- Logits/rejected: -2.1437
- Logits/chosen: -2.1850
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: 1e-05
- train_batch_size: 2
- eval_batch_size: 1
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 4
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 100
- training_steps: 1000
Training results
| Training Loss | Epoch | Step | Validation Loss | Rewards/chosen | Rewards/rejected | Rewards/accuracies | Rewards/margins | Logps/rejected | Logps/chosen | Logits/rejected | Logits/chosen |
|---|---|---|---|---|---|---|---|---|---|---|---|
| 0.0 | 0.2667 | 50 | 0.0000 | 0.6961 | -17.4863 | 1.0 | 18.1824 | -129.1691 | -34.4066 | -2.1527 | -2.1951 |
| 0.0 | 0.5333 | 100 | 0.0000 | 0.8907 | -18.1183 | 1.0 | 19.0091 | -131.2758 | -33.7579 | -2.1529 | -2.1947 |
| 0.0 | 0.8 | 150 | 0.0000 | 0.9820 | -18.5294 | 1.0 | 19.5115 | -132.6462 | -33.4536 | -2.1513 | -2.1928 |
| 0.0 | 1.0667 | 200 | 0.0000 | 1.0722 | -18.8065 | 1.0 | 19.8787 | -133.5699 | -33.1530 | -2.1495 | -2.1912 |
| 0.0 | 1.3333 | 250 | 0.0000 | 1.1163 | -19.0003 | 1.0 | 20.1166 | -134.2158 | -33.0060 | -2.1500 | -2.1915 |
| 0.0 | 1.6 | 300 | 0.0000 | 1.1575 | -19.1221 | 1.0 | 20.2797 | -134.6219 | -32.8687 | -2.1487 | -2.1902 |
| 0.0 | 1.8667 | 350 | 0.0000 | 1.1936 | -19.2727 | 1.0 | 20.4663 | -135.1237 | -32.7483 | -2.1484 | -2.1898 |
| 0.0 | 2.1333 | 400 | 0.0000 | 1.2250 | -19.4013 | 1.0 | 20.6263 | -135.5525 | -32.6439 | -2.1470 | -2.1883 |
| 0.0 | 2.4 | 450 | 0.0000 | 1.2436 | -19.5065 | 1.0 | 20.7501 | -135.9030 | -32.5818 | -2.1471 | -2.1884 |
| 0.0 | 2.6667 | 500 | 0.0000 | 1.2727 | -19.5732 | 1.0 | 20.8459 | -136.1255 | -32.4849 | -2.1464 | -2.1878 |
| 0.0 | 2.9333 | 550 | 0.0000 | 1.2747 | -19.6343 | 1.0 | 20.9090 | -136.3291 | -32.4779 | -2.1455 | -2.1868 |
| 0.0 | 3.2 | 600 | 0.0000 | 1.2898 | -19.6831 | 1.0 | 20.9729 | -136.4919 | -32.4277 | -2.1451 | -2.1865 |
| 0.0 | 3.4667 | 650 | 0.0000 | 1.2917 | -19.7102 | 1.0 | 21.0019 | -136.5822 | -32.4213 | -2.1441 | -2.1854 |
| 0.0 | 3.7333 | 700 | 0.0000 | 1.3059 | -19.7362 | 1.0 | 21.0421 | -136.6687 | -32.3740 | -2.1445 | -2.1859 |
| 0.0 | 4.0 | 750 | 0.0000 | 1.3080 | -19.7657 | 1.0 | 21.0737 | -136.7672 | -32.3670 | -2.1434 | -2.1848 |
| 0.0 | 4.2667 | 800 | 0.0000 | 1.3175 | -19.7744 | 1.0 | 21.0919 | -136.7961 | -32.3354 | -2.1443 | -2.1855 |
| 0.0 | 4.5333 | 850 | 0.0000 | 1.3043 | -19.7630 | 1.0 | 21.0674 | -136.7582 | -32.3793 | -2.1441 | -2.1855 |
| 0.0 | 4.8 | 900 | 0.0000 | 1.3061 | -19.7637 | 1.0 | 21.0698 | -136.7606 | -32.3735 | -2.1437 | -2.1850 |
| 0.0 | 5.0667 | 950 | 0.0000 | 1.3061 | -19.7637 | 1.0 | 21.0698 | -136.7606 | -32.3735 | -2.1437 | -2.1850 |
| 0.0 | 5.3333 | 1000 | 0.0000 | 1.3061 | -19.7637 | 1.0 | 21.0698 | -136.7606 | -32.3735 | -2.1437 | -2.1850 |
Framework versions
- Transformers 4.42.4
- Pytorch 2.0.0+cu117
- Datasets 2.20.0
- Tokenizers 0.19.1
- Task
- Text Generation
- Library
- transformers
- Type
- mistral
- License
- apache-2.0
- Created
- Jul 12, 2024
- Updated
- Jul 12, 2024