mihirinamdar/llama2-7b-bot-v1

Miniguanaco-7b Assistant Bot

This is a Miniguanaco-7b assistant bot, fine-tuned using QLoRA (4-bit precision) on the mlabonne/guanaco-llama2-1k dataset, a subset of timdettmers/openassistant-guanaco.

Training

The model was trained on a Google Colab notebook with a T4 GPU and high RAM. Please note that it is primarily designed for educational purposes and may not be optimized for production-level inference.

Usage

To use the Miniguanaco-7b assistant bot, you can follow the code example below:

# pip install transformers accelerate

from transformers import AutoTokenizer
import transformers
import torch

model = "mlabonne/llama-2-7b-miniguanaco"
prompt = "What is a large language model?"

tokenizer = AutoTokenizer.from_pretrained(model)

pipeline = transformers.pipeline(
    "text-generation",
    model=model,
    torch_dtype=torch.float16,
    device_map="auto",
)

sequences = pipeline(
    f'<s>[INST] {prompt} [/INST]',
    do_sample=True,
    top_k=10,
    num_return_sequences=1,
    eos_token_id=tokenizer.eos_token_id,
    max_length=200,
)

for seq in sequences:
    print(f"Result: {seq['generated_text']}")