|
|
--- |
|
|
library_name: peft |
|
|
base_model: mistralai/Mistral-7B-Instruct-v0.1 |
|
|
pipeline_tag: text2text-generation |
|
|
datasets: |
|
|
- bugdaryan/sql-create-context-instruction |
|
|
tags: |
|
|
- Mistral |
|
|
- PEFT |
|
|
- LoRA |
|
|
- SQL |
|
|
--- |
|
|
|
|
|
### Model Description |
|
|
|
|
|
<!-- Provide a longer summary of what this model is. --> |
|
|
SQL Generation model which is fine-tuned on the Mistral-7B-Instruct-v0.1. |
|
|
|
|
|
|
|
|
## 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: False |
|
|
- bnb_4bit_compute_dtype: float16 |
|
|
|
|
|
### Framework versions |
|
|
|
|
|
|
|
|
- PEFT 0.6.0.dev0 |
