Downloads · 30 days
2
5% of all-time downloads
Hyponatremia_M2_1000steps_1e6rate_05beta_DPO
tsavage68/Hyponatremia_M2_1000steps_1e6rate_05beta_DPO
Hyponatremia_M2_1000steps_1e6rate_05beta_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
44
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
- 66a29e29
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_1e6rate_05beta_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: 0.6008
- Rewards/rejected: -17.4429
- Rewards/accuracies: 1.0
- Rewards/margins: 18.0437
- Logps/rejected: -105.7672
- Logps/chosen: -35.5255
- Logits/rejected: -2.2010
- Logits/chosen: -2.2151
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-06
- 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 | 0.8519 | -13.5569 | 1.0 | 14.4088 | -97.9952 | -35.0232 | -2.2197 | -2.2286 |
| 0.0 | 0.5333 | 100 | 0.0000 | 0.7757 | -15.0466 | 1.0 | 15.8223 | -100.9746 | -35.1756 | -2.2145 | -2.2252 |
| 0.0 | 0.8 | 150 | 0.0000 | 0.7362 | -15.6205 | 1.0 | 16.3567 | -102.1225 | -35.2547 | -2.2112 | -2.2225 |
| 0.0 | 1.0667 | 200 | 0.0000 | 0.7126 | -16.0051 | 1.0 | 16.7178 | -102.8917 | -35.3018 | -2.2104 | -2.2222 |
| 0.0 | 1.3333 | 250 | 0.0000 | 0.7055 | -16.3418 | 1.0 | 17.0472 | -103.5650 | -35.3161 | -2.2085 | -2.2207 |
| 0.0 | 1.6 | 300 | 0.0000 | 0.6729 | -16.6069 | 1.0 | 17.2798 | -104.0952 | -35.3812 | -2.2065 | -2.2194 |
| 0.0 | 1.8667 | 350 | 0.0000 | 0.6486 | -16.8048 | 1.0 | 17.4534 | -104.4910 | -35.4299 | -2.2060 | -2.2192 |
| 0.0 | 2.1333 | 400 | 0.0000 | 0.6223 | -17.0030 | 1.0 | 17.6253 | -104.8875 | -35.4825 | -2.2043 | -2.2178 |
| 0.0 | 2.4 | 450 | 0.0000 | 0.6288 | -17.1153 | 1.0 | 17.7441 | -105.1121 | -35.4695 | -2.2031 | -2.2168 |
| 0.0 | 2.6667 | 500 | 0.0000 | 0.5934 | -17.2319 | 1.0 | 17.8253 | -105.3453 | -35.5403 | -2.2031 | -2.2170 |
| 0.0 | 2.9333 | 550 | 0.0000 | 0.5881 | -17.3028 | 1.0 | 17.8910 | -105.4871 | -35.5508 | -2.2022 | -2.2163 |
| 0.0 | 3.2 | 600 | 0.0000 | 0.5817 | -17.3752 | 1.0 | 17.9569 | -105.6318 | -35.5636 | -2.2020 | -2.2161 |
| 0.0 | 3.4667 | 650 | 0.0000 | 0.5838 | -17.3847 | 1.0 | 17.9685 | -105.6509 | -35.5594 | -2.2014 | -2.2156 |
| 0.0 | 3.7333 | 700 | 0.0000 | 0.5981 | -17.4078 | 1.0 | 18.0059 | -105.6971 | -35.5308 | -2.2015 | -2.2156 |
| 0.0 | 4.0 | 750 | 0.0000 | 0.5820 | -17.4428 | 1.0 | 18.0248 | -105.7672 | -35.5631 | -2.2015 | -2.2157 |
| 0.0 | 4.2667 | 800 | 0.0000 | 0.5894 | -17.4274 | 1.0 | 18.0168 | -105.7363 | -35.5482 | -2.2019 | -2.2159 |
| 0.0 | 4.5333 | 850 | 0.0000 | 0.6045 | -17.4195 | 1.0 | 18.0239 | -105.7204 | -35.5181 | -2.2013 | -2.2155 |
| 0.0 | 4.8 | 900 | 0.0000 | 0.6008 | -17.4429 | 1.0 | 18.0437 | -105.7672 | -35.5255 | -2.2010 | -2.2151 |
| 0.0 | 5.0667 | 950 | 0.0000 | 0.6008 | -17.4429 | 1.0 | 18.0437 | -105.7672 | -35.5255 | -2.2010 | -2.2151 |
| 0.0 | 5.3333 | 1000 | 0.0000 | 0.6008 | -17.4429 | 1.0 | 18.0437 | -105.7672 | -35.5255 | -2.2010 | -2.2151 |
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