---
library_name: transformers
license: apache-2.0
datasets:
- Technotech/sd-prompt-instruct
language:
- en
tags:
- stable-diffusion
- instruct
- magic-prompt
- natural language inference
---
# Stable Diffusion Prompt Instruct 3B (OpenLlama v2 3B)

Trained for 0.4 epochs (test) on [Technotech/sd-prompt-instruct](https://huggingface.co/datasets/Technotech/sd-prompt-instruct).

## Prompt Format

```
### Instruction: {prompt}
### Response: {response}
```

## Training procedure


The following `bitsandbytes` quantization config was used during training:
- 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: fp4
- bnb_4bit_use_double_quant: False
- bnb_4bit_compute_dtype: float32
### Framework versions


- PEFT 0.5.0.dev0