train_2024-04-14-12-24-44
This model is a fine-tuned version of Qwen/Qwen1.5-1.8B on the CMBanswermerge, the belle0.5Mrandom275220 and the HuatuoGPT2sftinstructGPT4 datasets.
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: 7e-05
- trainbatchsize: 4
- evalbatchsize: 8
- seed: 42
- gradientaccumulationsteps: 8
- totaltrainbatch_size: 32
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lrschedulertype: cosine
- lrschedulerwarmup_steps: 20
- num_epochs: 2.0
- mixedprecisiontraining: Native AMP
Training results
Framework versions
- Transformers 4.39.1
- Pytorch 2.2.0+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2
评论