README.md

---
library_name: peft
---
## Txt2Img LoRA for HuggingFaceH4_zephyr-7b-alpha

Use the lora with HuggingFaceH4_zephyr-7b-alpha model. I use it with the oobabooga's chat API in instruct mode.

Examples:

![image/png](https://cdn-uploads.huggingface.co/production/uploads/639c4f3b1c597f893b691751/Y3G5KAALNBYDz-d_MyhWZ.png)

## Training procedure


The following `bitsandbytes` quantization config was used during training:
- quant_method: QuantizationMethod.BITS_AND_BYTES
- 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

The following `bitsandbytes` quantization config was used during training:
- quant_method: QuantizationMethod.BITS_AND_BYTES
- 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 0.5.0

- PEFT 0.5.0