Downloads · 30 days
2
5% of all-time downloads
Hyponatremia_M2_1000steps_1e7rate_05beta_DPO
tsavage68/Hyponatremia_M2_1000steps_1e7rate_05beta_DPO
Hyponatremia_M2_1000steps_1e7rate_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
43
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
- 3e144b73
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_1e7rate_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: 0.6394
- Rewards/rejected: -14.1952
- Rewards/accuracies: 1.0
- Rewards/margins: 14.8346
- Logps/rejected: -99.2718
- Logps/chosen: -35.4482
- Logits/rejected: -2.2139
- Logits/chosen: -2.2236
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-07
- 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 |
|---|---|---|---|---|---|---|---|---|---|---|---|
| 0.0009 | 0.2667 | 50 | 0.0000 | 0.8288 | -9.6992 | 1.0 | 10.5280 | -90.2799 | -35.0694 | -2.2283 | -2.2332 |
| 0.0 | 0.5333 | 100 | 0.0000 | 0.7731 | -11.4586 | 1.0 | 12.2317 | -93.7986 | -35.1808 | -2.2236 | -2.2301 |
| 0.0 | 0.8 | 150 | 0.0000 | 0.7528 | -12.1438 | 1.0 | 12.8966 | -95.1690 | -35.2215 | -2.2211 | -2.2282 |
| 0.0 | 1.0667 | 200 | 0.0000 | 0.7382 | -12.5986 | 1.0 | 13.3368 | -96.0787 | -35.2507 | -2.2202 | -2.2278 |
| 0.0 | 1.3333 | 250 | 0.0000 | 0.7142 | -13.0630 | 1.0 | 13.7772 | -97.0075 | -35.2987 | -2.2180 | -2.2262 |
| 0.0 | 1.6 | 300 | 0.0000 | 0.6864 | -13.3376 | 1.0 | 14.0240 | -97.5566 | -35.3542 | -2.2171 | -2.2257 |
| 0.0 | 1.8667 | 350 | 0.0000 | 0.6758 | -13.5516 | 1.0 | 14.2274 | -97.9847 | -35.3755 | -2.2162 | -2.2251 |
| 0.0 | 2.1333 | 400 | 0.0000 | 0.6502 | -13.7675 | 1.0 | 14.4177 | -98.4164 | -35.4266 | -2.2153 | -2.2243 |
| 0.0 | 2.4 | 450 | 0.0000 | 0.6422 | -13.9291 | 1.0 | 14.5712 | -98.7396 | -35.4427 | -2.2146 | -2.2239 |
| 0.0 | 2.6667 | 500 | 0.0000 | 0.6435 | -14.0130 | 1.0 | 14.6564 | -98.9074 | -35.4401 | -2.2144 | -2.2239 |
| 0.0 | 2.9333 | 550 | 0.0000 | 0.6341 | -14.0963 | 1.0 | 14.7304 | -99.0741 | -35.4589 | -2.2145 | -2.2241 |
| 0.0 | 3.2 | 600 | 0.0000 | 0.6375 | -14.1392 | 1.0 | 14.7767 | -99.1599 | -35.4521 | -2.2144 | -2.2239 |
| 0.0 | 3.4667 | 650 | 0.0000 | 0.6270 | -14.1769 | 1.0 | 14.8039 | -99.2353 | -35.4731 | -2.2140 | -2.2237 |
| 0.0 | 3.7333 | 700 | 0.0000 | 0.6401 | -14.1999 | 1.0 | 14.8400 | -99.2813 | -35.4469 | -2.2141 | -2.2238 |
| 0.0 | 4.0 | 750 | 0.0000 | 0.6332 | -14.1728 | 1.0 | 14.8060 | -99.2270 | -35.4606 | -2.2135 | -2.2232 |
| 0.0 | 4.2667 | 800 | 0.0000 | 0.6216 | -14.2039 | 1.0 | 14.8255 | -99.2892 | -35.4838 | -2.2142 | -2.2239 |
| 0.0 | 4.5333 | 850 | 0.0000 | 0.6349 | -14.1877 | 1.0 | 14.8226 | -99.2568 | -35.4572 | -2.2140 | -2.2237 |
| 0.0 | 4.8 | 900 | 0.0000 | 0.6394 | -14.1952 | 1.0 | 14.8346 | -99.2718 | -35.4482 | -2.2139 | -2.2236 |
| 0.0 | 5.0667 | 950 | 0.0000 | 0.6394 | -14.1952 | 1.0 | 14.8346 | -99.2718 | -35.4482 | -2.2139 | -2.2236 |
| 0.0 | 5.3333 | 1000 | 0.0000 | 0.6394 | -14.1952 | 1.0 | 14.8346 | -99.2718 | -35.4482 | -2.2139 | -2.2236 |
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