Downloads · 30 days
3
11% of all-time downloads
model_hh_shp2_200
guoyu-zhang/model_hh_shp2_200
model_hh_shp2_200 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. The card lists the license as llama2.
should probably proofread and complete it, then remove this comment. --
All-time downloads
28
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
- License
- llama2
- Access
- Public
- Created
- Apr 17, 2024
- Updated
- Apr 17, 2024
- SHA
- bf14a029
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_shp2_200
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: 2.0980
- Rewards/chosen: -2.6010
- Rewards/rejected: -3.0421
- Rewards/accuracies: 0.5700
- Rewards/margins: 0.4411
- Logps/rejected: -224.1422
- Logps/chosen: -244.7963
- Logits/rejected: -0.4942
- Logits/chosen: -0.5683
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 |
|---|---|---|---|---|---|---|---|---|---|---|---|
| 0.0 | 8.0 | 100 | 2.1281 | -2.3340 | -2.7556 | 0.5700 | 0.4216 | -223.8238 | -244.4996 | -0.4932 | -0.5692 |
| 0.0 | 16.0 | 200 | 2.0805 | -2.4076 | -2.9037 | 0.5800 | 0.4961 | -223.9884 | -244.5814 | -0.4930 | -0.5685 |
| 0.0 | 24.0 | 300 | 2.1050 | -2.5116 | -2.9546 | 0.5600 | 0.4430 | -224.0449 | -244.6970 | -0.4939 | -0.5685 |
| 0.0 | 32.0 | 400 | 2.1003 | -2.5211 | -2.9879 | 0.5600 | 0.4667 | -224.0819 | -244.7076 | -0.4938 | -0.5683 |
| 0.0 | 40.0 | 500 | 2.1098 | -2.5733 | -3.0310 | 0.5700 | 0.4576 | -224.1297 | -244.7656 | -0.4935 | -0.5677 |
| 0.0 | 48.0 | 600 | 2.0969 | -2.5725 | -3.0456 | 0.5700 | 0.4731 | -224.1461 | -244.7647 | -0.4939 | -0.5680 |
| 0.0 | 56.0 | 700 | 2.1051 | -2.6073 | -3.0413 | 0.5500 | 0.4341 | -224.1413 | -244.8033 | -0.4936 | -0.5679 |
| 0.0 | 64.0 | 800 | 2.0586 | -2.5796 | -3.0722 | 0.5600 | 0.4926 | -224.1756 | -244.7725 | -0.4935 | -0.5680 |
| 0.0 | 72.0 | 900 | 2.1077 | -2.5920 | -3.0537 | 0.5700 | 0.4617 | -224.1551 | -244.7863 | -0.4936 | -0.5682 |
| 0.0 | 80.0 | 1000 | 2.0980 | -2.6010 | -3.0421 | 0.5700 | 0.4411 | -224.1422 | -244.7963 | -0.4942 | -0.5683 |
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
- License
- llama2
- Created
- Apr 17, 2024
- Updated
- Apr 17, 2024