• Developed by: Mahmoud Ibrahim

  • How to use :

! pip install transformers bitsandbytes
from transformers import AutoTokenizer, AutoModelForCausalLM
from IPython.display import Markdown
import textwrap 

# Load tokenizer and model
tokenizer = AutoTokenizer.from_pretrained("MahmoudIbrahim/Mistral_12b_Arabic")
model = AutoModelForCausalLM.from_pretrained("MahmoudIbrahim/Mistral_12b_Arabic",load_in_4bit =True)

alpaca_prompt = """ููŠู…ุง ูŠู„ูŠ ุชุนู„ูŠู…ุงุช ุชุตู ู…ู‡ู…ุฉุŒ ุฅู„ู‰ ุฌุงู†ุจ ู…ุฏุฎู„ ูŠูˆูุฑ ุณูŠุงู‚ุงู‹ ุฅุถุงููŠุงู‹. ุงูƒุชุจ ุงุณุชุฌุงุจุฉ ุชููƒู…ู„ ุงู„ุทู„ุจ ุจุดูƒู„ ู…ู†ุงุณุจ.

### ุงู„ุชุนู„ูŠู…ุงุช:
{}

### ุงู„ุงุณุชุฌุงุจุฉ:
{}"""

# Format the prompt with instruction and an empty output placeholder
formatted_prompt = alpaca_prompt.format(
    "ูƒูŠู ูŠู…ูƒู† ู„ู„ุญูƒูˆู…ุฉ ุงู„ู…ุตุฑูŠุฉ ูˆุงู„ู…ุฌุชู…ุน ูƒูƒู„ ุฃู† ูŠุนุฒุฒูˆุง ู…ู† ู‚ุฏุฑุฉ ุงู„ุจู„ุงุฏ ุนู„ู‰ ุชุญู‚ูŠู‚ ุงู„ุชู†ู…ูŠุฉ ุงู„ู…ุณุชุฏุงู…ุฉุŸ "   ,  # instruction
    ""  # Leave output blank for generation
)

# Tokenize the formatted string directly
input_ids = tokenizer.encode(formatted_prompt, return_tensors="pt")  # Use 'cuda' if you want to run on GPU

def to_markdown(text):
    text = text.replace('โ€ข','*')
    return Markdown(textwrap.indent(text, '>', predicate=lambda _: True))

# Generate text
output = model.generate(
    input_ids,
    max_length=128,            # Adjust max length as needed
    num_return_sequences=1,     # Number of generated responses
    no_repeat_ngram_size=2,     # Prevent repetition
    top_k=50,                   # Filter to top-k tokens
    top_p=0.9,                  # Use nucleus sampling
    temperature=0.7 ,            # Control creativity level
  
)

generated_text = tokenizer.decode(output[0], skip_special_tokens=True)
to_markdown(generated_text)

The model response :

image/png

Downloads last month
34
Safetensors
Model size
6.97B params
Tensor type
F32
ยท
FP16
ยท
U8
ยท
Inference Examples
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated) instead.

Model tree for MahmoudIbrahim/Mistral_Nemo_Arabic

Quantized
(13)
this model

Dataset used to train MahmoudIbrahim/Mistral_Nemo_Arabic