LoRA Text Generation Model

This repository contains a fine-tuned LoRA model designed for text generation.

Training Details

Fine-Tuning Procedure

The model was fine-tuned using the following bitsandbytes quantization configuration:

• 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: bfloat16

Framework Versions

• PEFT 0.4.0

How to Use

Installation

Install the necessary libraries:

pip install transformers