|
--- |
|
library_name: peft |
|
base_model: meta-llama/Llama-2-13b-hf |
|
datasets: |
|
- harpomaxx/jurisgpt |
|
--- |
|
|
|
## Training Details |
|
|
|
The training script is available [here](https://github.com/harpomaxx/JurisGPT/blob/e15b844875c3f7e27eaf1a0e5dc13f069e38fee0/code/python/scripts/finetune.py) |
|
|
|
### Training Data |
|
|
|
<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. --> |
|
|
|
## Training procedure |
|
|
|
|
|
The following `bitsandbytes` quantization config was used during training: |
|
- quant_method: bitsandbytes |
|
- load_in_8bit: False |
|
- load_in_4bit: True |
|
- llm_int8_threshold: 6.0 |
|
- llm_int8_skip_modules: None |
|
- llm_int8_enable_fp32_cpu_offload: False |
|
- llm_int8_has_fp16_weight: False |
|
- bnb_4bit_quant_type: nf4 |
|
- bnb_4bit_use_double_quant: True |
|
- bnb_4bit_compute_dtype: float16 |
|
|
|
### Framework versions |
|
|
|
- PEFT 0.6.2 |
