|
--- |
|
license: apache-2.0 |
|
--- |
|
|
|
## Model Description |
|
|
|
Master is a collection of LLMs trained using human-collected seed questions and regenerate the answers with a mixture of high performance Open-source LLMs. |
|
|
|
**Master-Yi-9B** is trained using the ORPO techniques. The model shows strong abilities in reasoning on coding and math questions. |
|
|
|
|
|
 |
|
|
|
## Prompt Template |
|
|
|
``` |
|
<|im_start|>system |
|
You are a helpful AI assistant.<|im_end|> |
|
<|im_start|>user |
|
What is the meaning of life?<|im_end|> |
|
<|im_start|>assistant |
|
``` |
|
|
|
## Training procedure |
|
|
|
### Training hyperparameters |
|
|
|
The following hyperparameters were used during training: |
|
- learning_rate: 2e-06 |
|
- train_batch_size: 2 |
|
- eval_batch_size: 2 |
|
- seed: 42 |
|
- distributed_type: multi-GPU |
|
- num_devices: 4 |
|
- gradient_accumulation_steps: 4 |
|
- total_train_batch_size: 32 |
|
- total_eval_batch_size: 8 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: cosine |
|
- lr_scheduler_warmup_steps: 100 |
|
- num_epochs: 1 |
|
|
|
### Training results |
|
|
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.34.1 |
|
- Pytorch 2.0.1+cu118 |
|
- Datasets 2.14.6 |
|
- Tokenizers 0.14.1 |
