Downloads · 30 days
4
15% of all-time downloads
model_shp2_dpo9
guoyu-zhang/model_shp2_dpo9
model_shp2_dpo9 is a machine learning model from guoyu-zhang. Use it for the machine learning task on the model card, and read the license before you ship it in a product. It is set up for peft.
should probably proofread and complete it, then remove this comment. --
All-time downloads
27
Public
Repo size
12.6 MB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors25.2 MB · 93%
At a glance
- Library
- peft
- Access
- Public
- Created
- Apr 17, 2024
- Updated
- Apr 17, 2024
- SHA
- 0bc2106f
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. -->
model_shp2_dpo9
This model is a fine-tuned version of meta-llama/Llama-2-7b-chat-hf on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 3.2776
- Rewards/chosen: -1.1246
- Rewards/rejected: -1.9364
- Rewards/accuracies: 0.4900
- Rewards/margins: 0.8117
- Logps/rejected: -216.7705
- Logps/chosen: -231.7811
- Logits/rejected: -0.9117
- Logits/chosen: -0.9723
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: 0.0005
- train_batch_size: 4
- eval_batch_size: 1
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 16
- 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 |
|---|---|---|---|---|---|---|---|---|---|---|---|
| 0.0604 | 2.67 | 100 | 1.8181 | 7.7452 | 7.8839 | 0.4900 | -0.1387 | -205.8591 | -221.9257 | -0.8827 | -0.9016 |
| 0.0268 | 5.33 | 200 | 2.9688 | -2.8174 | -2.9847 | 0.4800 | 0.1673 | -217.9353 | -233.6619 | -1.1131 | -1.1645 |
| 0.0069 | 8.0 | 300 | 3.0520 | 6.6739 | 6.0279 | 0.5600 | 0.6459 | -207.9212 | -223.1161 | -0.9294 | -1.0100 |
| 0.0 | 10.67 | 400 | 3.2909 | -1.1251 | -1.8955 | 0.4900 | 0.7704 | -216.7250 | -231.7816 | -0.9109 | -0.9719 |
| 0.0 | 13.33 | 500 | 3.2845 | -1.1008 | -1.9104 | 0.5 | 0.8096 | -216.7416 | -231.7545 | -0.9109 | -0.9718 |
| 0.0 | 16.0 | 600 | 3.3090 | -1.1249 | -1.9231 | 0.4900 | 0.7983 | -216.7558 | -231.7813 | -0.9112 | -0.9722 |
| 0.0 | 18.67 | 700 | 3.2953 | -1.1118 | -1.9182 | 0.4900 | 0.8063 | -216.7503 | -231.7668 | -0.9116 | -0.9723 |
| 0.0 | 21.33 | 800 | 3.2821 | -1.1048 | -1.9227 | 0.4900 | 0.8179 | -216.7553 | -231.7590 | -0.9116 | -0.9726 |
| 0.0 | 24.0 | 900 | 3.2731 | -1.1170 | -1.9616 | 0.4900 | 0.8445 | -216.7985 | -231.7726 | -0.9111 | -0.9723 |
| 0.0 | 26.67 | 1000 | 3.2776 | -1.1246 | -1.9364 | 0.4900 | 0.8117 | -216.7705 | -231.7811 | -0.9117 | -0.9723 |
Framework versions
- PEFT 0.10.0
- Transformers 4.39.1
- Pytorch 2.2.1+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2
- Library
- peft
- Created
- Apr 17, 2024
- Updated
- Apr 17, 2024