---
library_name: peft
base_model: Trelis/Llama-2-7b-chat-hf-function-calling-v2
datasets:
- BI55/MedText
pipeline_tag: text-generation
---

# Model still under developing


## How to Get Started with the Model

https://colab.research.google.com/drive/1eFzHy6eARkhgy4Uxyre5KsdRbQII50jx?usp=sharing


# GGUF version 
https://huggingface.co./Elkhayyat17/llama2medical-GGUF
# Merged version 
https://huggingface.co./Elkhayyat17/merge-PEFT-Llama-2-7b-MedText

## 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 0.6.0.dev0