KillerShoaib/llama-3-8b-bangla-lora

LLama-3 Bangla LoRA

• Developed by: KillerShoaib
• License: apache-2.0
• Finetuned from model : unsloth/llama-3-8b-bnb-4bit
• Datset used for fine-tuning : iamshnoo/alpaca-cleaned-bengali

LoRA Adapter

This is not the entire model, but rather only the LoRA adapter.

Llama-3 Bangla Different Formats

• 4-bit quantized(QLoRA) - KillerShoaib/llama-3-8b-bangla-4bit
• GGUF q4_k_m - KillerShoaib/llama-3-8b-bangla-GGUF-Q4_K_M

Model Details

Llama 3 8 billion model was finetuned using unsloth package on a cleaned Bangla alpaca dataset. The model is finetuned for 2 epoch on a single T4 GPU.

Pros & Cons of the Model

Pros

• The model can comprehend the Bangla language, including its semantic nuances
• Given context model can answer the question based on the context

Cons

• Model is unable to do creative or complex work. i.e: creating a poem or solving a math problem in Bangla
• Since the size of the dataset was small, the model lacks lot of general knowledge in Bangla

Run The Model

FastLanguageModel from unsloth for 2x faster inference

from unsloth import FastLanguageModel
model, tokenizer = FastLanguageModel.from_pretrained(
    model_name = "KillerShoaib/llama-3-8b-bangla-lora",
    max_seq_length = 2048,
    dtype = None,
    load_in_4bit = True,
)
FastLanguageModel.for_inference(model)

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

### Instruction:
{}

### Input:
{}

### Response:
{}"""

# input with instruction and input
inputs = tokenizer(
[
    alpaca_prompt.format(
        "সুস্থ থাকার তিনটি উপায় বলুন", # instruction
        "", # input
        "", # output - leave this blank for generation!
    )
], return_tensors = "pt").to("cuda")

# generating the output and decoding it
outputs = model.generate(**inputs, max_new_tokens = 2048, use_cache = True)
tokenizer.batch_decode(outputs)

AutoModelForPeftCausalLM from Hugginface

from peft import AutoPeftModelForCausalLM
from transformers import AutoTokenizer
model = AutoPeftModelForCausalLM.from_pretrained(
    "KillerShoaib/llama-3-8b-bangla-lora",
    load_in_4bit = True,
)
tokenizer = AutoTokenizer.from_pretrained("KillerShoaib/llama-3-8b-bangla-lora")

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

### Instruction:
{}

### Input:
{}

### Response:
{}"""

inputs = tokenizer(
[
    alpaca_prompt.format(
        "সুস্থ থাকার তিনটি উপায় বলুন", # instruction
        "", # input
        "", # output - leave this blank for generation!
    )
], return_tensors = "pt").to("cuda")

outputs = model.generate(**inputs, max_new_tokens = 1024, use_cache = True)
tokenizer.batch_decode(outputs)

AutoModelForPeftCausalLM can be hopelessly slow, since 4bit model downloading is not supported. Use this only if you don't have unsloth installed

Inference Script & Github Repo

• Google Colab - Llama-3 8b Bangla Inference Script
• Github Repo - Llama-3 Bangla