Downloads · 30 days
4
12% of all-time downloads
model_hh_shp1_dpo7
guoyu-zhang/model_hh_shp1_dpo7
model_hh_shp1_dpo7 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
33
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 16, 2024
- Updated
- Apr 16, 2024
- SHA
- 01684737
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_hh_shp1_dpo7
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: 2.8000
- Rewards/chosen: -7.0133
- Rewards/rejected: -7.1903
- Rewards/accuracies: 0.4800
- Rewards/margins: 0.1770
- Logps/rejected: -240.3288
- Logps/chosen: -247.0569
- Logits/rejected: -1.1073
- Logits/chosen: -1.0850
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 |
|---|---|---|---|---|---|---|---|---|---|---|---|
| 0.0959 | 2.67 | 100 | 1.3530 | -3.1387 | -3.5029 | 0.5500 | 0.3642 | -235.0611 | -241.5217 | -0.6924 | -0.6728 |
| 0.0119 | 5.33 | 200 | 2.3339 | -9.2972 | -9.3321 | 0.5400 | 0.0349 | -243.3885 | -250.3196 | -0.7541 | -0.7327 |
| 0.063 | 8.0 | 300 | 3.1443 | -10.1013 | -9.9537 | 0.4800 | -0.1475 | -244.2765 | -251.4682 | -0.8796 | -0.8535 |
| 0.0016 | 10.67 | 400 | 2.8783 | -10.1491 | -10.5391 | 0.5100 | 0.3900 | -245.1127 | -251.5365 | -1.0090 | -0.9845 |
| 0.0 | 13.33 | 500 | 2.8233 | -7.0398 | -7.1738 | 0.4700 | 0.1340 | -240.3052 | -247.0948 | -1.1069 | -1.0847 |
| 0.0 | 16.0 | 600 | 2.7833 | -7.0362 | -7.2095 | 0.4800 | 0.1733 | -240.3561 | -247.0895 | -1.1075 | -1.0848 |
| 0.0 | 18.67 | 700 | 2.7984 | -7.0122 | -7.1531 | 0.4800 | 0.1409 | -240.2756 | -247.0552 | -1.1069 | -1.0844 |
| 0.0 | 21.33 | 800 | 2.7969 | -7.0208 | -7.1923 | 0.4800 | 0.1714 | -240.3316 | -247.0676 | -1.1074 | -1.0849 |
| 0.0 | 24.0 | 900 | 2.8036 | -7.0360 | -7.1975 | 0.4800 | 0.1615 | -240.3390 | -247.0892 | -1.1075 | -1.0850 |
| 0.0 | 26.67 | 1000 | 2.8000 | -7.0133 | -7.1903 | 0.4800 | 0.1770 | -240.3288 | -247.0569 | -1.1073 | -1.0850 |
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 16, 2024
- Updated
- Apr 16, 2024