A suitable name for this section could be:

Model Description

This model is fine-tuned from LLama 3.1 8B, enhanced for improved capability in the Arabic language. It was fine-tuned on 10,000 samples using Alpaca prompt instructions.

Please refer to this repository when using the model.

To perform inference using these LoRA adapters, please use the following code:

# Installs Unsloth, Xformers (Flash Attention) and all other packages!
!pip install "unsloth[colab-new] @ git+https://github.com/unslothai/unsloth.git"
!pip install --no-deps "xformers<0.0.27" "trl<0.9.0" peft accelerate bitsandbytes

from unsloth import FastLanguageModel
model, tokenizer = FastLanguageModel.from_pretrained(
    model_name = "AhmedBou/Arabic-Meta-Llama-3.1-8B_LoRA", # YOUR MODEL YOU USED FOR TRAINING
    max_seq_length = 2048,
    dtype = None,
    load_in_4bit = True,
)
FastLanguageModel.for_inference(model) # Enable native 2x faster inference

alpaca_prompt = """Below is an instruction that describes a task, paired with an input that provides further context. Write a response that appropriately completes the request.

### Instruction:
{}

### Input:
{}

### Response:
{}"""

inputs = tokenizer(
[
    alpaca_prompt.format(
        "قم بصياغة الجملة الإنجليزية التالية باللغة العربية.", # instruction
        "We hope that the last cases will soon be resolved through the mechanisms established for this purpose.", # input
        "", # output - leave this blank for generation!
    )
], return_tensors = "pt").to("cuda")

from transformers import TextStreamer
text_streamer = TextStreamer(tokenizer)
_ = model.generate(**inputs, streamer = text_streamer, max_new_tokens = 128)

The Outout is:

<|begin_of_text|>Below is an instruction that describes a task, paired with an input that provides further context. Write a response that appropriately completes the request.

### Instruction:
قم بصياغة الجملة الإنجليزية التالية باللغة العربية.

### Input:
We hope that the last cases will soon be resolved through the mechanisms established for this purpose.

### Response:
وأملنا في أن يكون هناك حل سريع للمواد الأخيرة من خلال الآليات المحددة لهذا الغرض.<|end_of_text|>

Uploaded model

• Developed by: AhmedBou
• License: apache-2.0
• Finetuned from model : unsloth/meta-llama-3.1-8b-bnb-4bit

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