Downloads · 30 days
2
4% of all-time downloads
Hyponatremia_M2_1000steps_1e5rate_01beta_DPO
tsavage68/Hyponatremia_M2_1000steps_1e5rate_01beta_DPO
Hyponatremia_M2_1000steps_1e5rate_01beta_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
47
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 11, 2024
- Updated
- Jul 11, 2024
- SHA
- 49e39f2b
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_01beta_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: 1.2112
- Rewards/rejected: -16.8668
- Rewards/accuracies: 1.0
- Rewards/margins: 18.0780
- Logps/rejected: -239.5494
- Logps/chosen: -24.6150
- Logits/rejected: -1.9501
- Logits/chosen: -2.0352
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 | 0.5819 | -15.3818 | 1.0 | 15.9637 | -224.6993 | -30.9082 | -1.9029 | -1.9913 |
| 0.0 | 0.5333 | 100 | 0.0000 | 0.8042 | -15.7156 | 1.0 | 16.5197 | -228.0370 | -28.6852 | -1.9187 | -2.0063 |
| 0.0 | 0.8 | 150 | 0.0000 | 0.9087 | -15.9319 | 1.0 | 16.8406 | -230.2003 | -27.6401 | -1.9294 | -2.0161 |
| 0.0 | 1.0667 | 200 | 0.0000 | 0.9784 | -16.0886 | 1.0 | 17.0670 | -231.7670 | -26.9426 | -1.9370 | -2.0230 |
| 0.0 | 1.3333 | 250 | 0.0000 | 1.0253 | -16.2250 | 1.0 | 17.2502 | -233.1312 | -26.4744 | -1.9412 | -2.0268 |
| 0.0 | 1.6 | 300 | 0.0000 | 1.0626 | -16.3349 | 1.0 | 17.3975 | -234.2300 | -26.1009 | -1.9459 | -2.0311 |
| 0.0 | 1.8667 | 350 | 0.0000 | 1.0970 | -16.4438 | 1.0 | 17.5408 | -235.3197 | -25.7575 | -1.9469 | -2.0321 |
| 0.0 | 2.1333 | 400 | 0.0000 | 1.1248 | -16.5266 | 1.0 | 17.6514 | -236.1472 | -25.4792 | -1.9492 | -2.0342 |
| 0.0 | 2.4 | 450 | 0.0000 | 1.1470 | -16.6094 | 1.0 | 17.7564 | -236.9759 | -25.2573 | -1.9495 | -2.0342 |
| 0.0 | 2.6667 | 500 | 0.0000 | 1.1677 | -16.6759 | 1.0 | 17.8435 | -237.6400 | -25.0502 | -1.9502 | -2.0352 |
| 0.0 | 2.9333 | 550 | 0.0000 | 1.1758 | -16.7249 | 1.0 | 17.9007 | -238.1303 | -24.9689 | -1.9507 | -2.0354 |
| 0.0 | 3.2 | 600 | 0.0000 | 1.1917 | -16.7740 | 1.0 | 17.9657 | -238.6219 | -24.8102 | -1.9492 | -2.0343 |
| 0.0 | 3.4667 | 650 | 0.0000 | 1.1987 | -16.8055 | 1.0 | 18.0042 | -238.9364 | -24.7404 | -1.9506 | -2.0357 |
| 0.0 | 3.7333 | 700 | 0.0000 | 1.2018 | -16.8329 | 1.0 | 18.0347 | -239.2105 | -24.7090 | -1.9508 | -2.0358 |
| 0.0 | 4.0 | 750 | 0.0000 | 1.2073 | -16.8544 | 1.0 | 18.0617 | -239.4252 | -24.6539 | -1.9517 | -2.0368 |
| 0.0 | 4.2667 | 800 | 0.0000 | 1.2085 | -16.8622 | 1.0 | 18.0707 | -239.5037 | -24.6419 | -1.9500 | -2.0351 |
| 0.0 | 4.5333 | 850 | 0.0000 | 1.2096 | -16.8665 | 1.0 | 18.0761 | -239.5468 | -24.6309 | -1.9514 | -2.0365 |
| 0.0 | 4.8 | 900 | 0.0000 | 1.2107 | -16.8691 | 1.0 | 18.0797 | -239.5722 | -24.6204 | -1.9506 | -2.0357 |
| 0.0 | 5.0667 | 950 | 0.0000 | 1.2112 | -16.8668 | 1.0 | 18.0780 | -239.5494 | -24.6150 | -1.9501 | -2.0352 |
| 0.0 | 5.3333 | 1000 | 0.0000 | 1.2112 | -16.8668 | 1.0 | 18.0780 | -239.5494 | -24.6150 | -1.9501 | -2.0352 |
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 11, 2024
- Updated
- Jul 11, 2024