metadata

base_model: inceptionai/jais-adapted-7b-chat
language:
  - ar
license: apache-2.0
tags:
  - text-generation-inference
  - transformers
  - unsloth
  - llama
  - trl
datasets:
  - Wajdi1976/Tunisian_Derja_Dataset
library_name: transformers

Model Overview

Labess is an open models instruction-tuned for Tunisian Derja, it's a continual pre-training version of jais-adapted-7b-chat with tunisian_Derja_Dataset

Uploaded model

Developed by: Wajdi1976
License: apache-2.0
Finetuned from model : inceptionai/jais-adapted-7b-chat

Usage

Below we share some code snippets on how to get quickly started with running the model. First, install the Transformers library with:

pip install unsloth

First, Load the Model

from unsloth import FastLanguageModel
import torch
max_seq_length = 600 # Choose any! We auto support RoPE Scaling internally!
dtype = None # None for auto detection. Float16 for Tesla T4, V100, Bfloat16 for Ampere+
load_in_4bit = True # Use 4bit quantization to reduce memory usage. Can be False.
model, tokenizer = FastLanguageModel.from_pretrained(
    model_name = "Wajdi1976/Labess",
    max_seq_length = max_seq_length,
    dtype = dtype,
    load_in_4bit = load_in_4bit,    
)

Second, Try the model

prompt_ar=" يمكنك الإجابة باللهجة التونسية فقط.\n\nأكمل المحادثة أدناه بين [|Human|] و [|AI|]:\n### Input: [|Human|] {Question}\n### Response: [|AI|]"
device = "cuda" if torch.cuda.is_available() else "cpu"
FastLanguageModel.for_inference(model)
if tokenizer.pad_token is None:
    tokenizer.pad_token = tokenizer.eos_token
def get_response(text, tokenizer=tokenizer, model=model):
    tokenized = tokenizer(text, return_tensors="pt")
    input_ids, attention_mask = tokenized['input_ids'].to(device), tokenized['attention_mask'].to(device)
    input_len = input_ids.shape[-1]
    generate_ids = model.generate(
        input_ids,
        attention_mask=attention_mask,
        top_p=1,
        temperature=0.3,
        max_length=600,
        min_length=input_len + 4,
        repetition_penalty=1.2,
        do_sample=True,
        pad_token_id=tokenizer.pad_token_id
    )
    response = tokenizer.batch_decode(
        generate_ids, skip_special_tokens=True, clean_up_tokenization_spaces=True
    )[0]
    response = response.split("### Response :")[-1].lstrip()
    return response

ques = "  آش نقصدو كي نقولو لاباس"
text = prompt_ar.format_map({'Question': ques})
print(get_response(text))

Response: لاباس تعني إنك بخير وما فما حتى مشكلة كيف تقول إنو كل شي باهي ومش مشكل نجم نعاونك في حاجة أخرى ولا عندك سؤال آخر تحب تسألوالإجابة هي هاكا يا صاحبيالدنيا ديما تكون خير كيما الواحد يكون مرتاح وباهي ويحس روحوا مع صحابو ومع الناس الكل هكة مهم برشا باش نكون عند حسن الظن ونتعامل بكل لطف ونكون طيب وحنون خاطر هذا يخلينا نواصلو علاقاتنا ونحمي صداقت

Citations

When using the Tunisian Derja Dataset dataset, please cite:

@model{linagora2025LLM-tn,
  author = {Wajdi Ghezaiel and Jean-Pierre Lorré},
  title = {Labess:Tunisian Derja LLM},
  year = {2025},
  month = {January},  
  url = {https://huggingface.co/datasets/Wajdi1976/Labess}
}

This llama model was trained 2x faster with Unsloth and Huggingface's TRL library.