Instructions to use nsomazr/blood-donation-llama32-3b-lora with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Local Apps Settings
- Unsloth Desktop
Blood Donation Advisory β Llama 3.2 3B (LoRA Adapters)
Bilingual (English + Kiswahili) blood donation advisor fine-tuned for Tanzania.
Evaluation Results β Llama 3.2 3B
NLP Quality (higher is better)
| Metric | Llama 3.2 1B | Llama 3.2 3B | Winner |
|---|---|---|---|
| ROUGE-1 | 0.4060 | 0.4410 | 3B β |
| ROUGE-2 | 0.1845 | 0.2639 | 3B β |
| ROUGE-L | 0.3039 | 0.3619 | 3B β |
| BLEU | 0.2134 | 0.2461 | 3B β |
| BERTScore Precision | 0.8741 | 0.8817 | 3B β |
| BERTScore Recall | 0.8628 | 0.8676 | 3B β |
| BERTScore F1 | 0.8683 | 0.8744 | 3B β |
| Perplexity β | 3.89 | 3.25 | 3B β |
Efficiency
| Metric | Llama 3.2 1B | Llama 3.2 3B |
|---|---|---|
| Training time (min) | 10.42 | 22.13 |
| Peak VRAM (GB) | 10.83 | 11.113 |
| LoRA VRAM (GB) | 9.738 | 8.361 |
System Prompt
You are a helpful blood donation advisor serving Tanzania.
Always respond in the same language the user writes in.
Quick Start (LoRA Adapters)
from unsloth import FastLanguageModel
model, tokenizer = FastLanguageModel.from_pretrained(
model_name="nsomazr/blood-donation-llama32-3b-lora",
max_seq_length=8192, dtype=None, load_in_4bit=True,
)
Inference Providers NEW
This model isn't deployed by any Inference Provider. π Ask for provider support
Model tree for nsomazr/blood-donation-llama32-3b-lora
Base model
meta-llama/Llama-3.2-3B-Instruct