File size: 2,630 Bytes
c88e2a4 651f3a2 6d32964 651f3a2 cdb21b9 6d32964 651f3a2 3e3905b 651f3a2 c88e2a4 651f3a2 381d701 92269f5 381d701 3e3905b 5e5fa62 651f3a2 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 |
import spaces
import gradio as gr
import torch
from transformers import AutoTokenizer, AutoModelForCausalLM
title = """# 🙋🏻♂️ Welcome to Tonic's Minitron-8B-Base"""
description = """
Minitron is a family of small language models (SLMs) obtained by pruning [NVIDIA's](https://huggingface.co/nvidia) Nemotron-4 15B model. We prune model embedding size, attention heads, and MLP intermediate dimension, following which, we perform continued training with distillation to arrive at the final models.
### Join us :
🌟TeamTonic🌟 is always making cool demos! Join our active builder's 🛠️community 👻 [](https://discord.gg/GWpVpekp) On 🤗Huggingface:[MultiTransformer](https://huggingface.co/MultiTransformer) On 🌐Github: [Tonic-AI](https://github.com/tonic-ai) & contribute to🌟 [BuildTonic](https://github.com/buildtonic/)🤗Big thanks to Yuvi Sharma and all the folks at huggingface for the community grant 🤗
"""
# Load the tokenizer and model
model_path = "nvidia/Minitron-8B-Base"
tokenizer = AutoTokenizer.from_pretrained(model_path)
device='cuda'
dtype=torch.bfloat16
model = AutoModelForCausalLM.from_pretrained(model_path, torch_dtype=dtype, device_map=device)
# Define the prompt format
def create_prompt(instruction):
PROMPT = '''Below is an instruction that describes a task.\n\nWrite a response that appropriately completes the request.\n\n### Instruction:\n{instruction}\n\n### Response:'''
return PROMPT.format(instruction=instruction)
@spaces.GPU
def respond(message, history, system_message, max_tokens, temperature, top_p):
prompt = create_prompt(message)
input_ids = tokenizer.encode(prompt, return_tensors="pt").to(model.device)
output_ids = model.generate(input_ids, max_length=50, num_return_sequences=1)
output_text = tokenizer.decode(output_ids[0], skip_special_tokens=True)
return output_text
demo = gr.ChatInterface(
title=gr.Markdown(title),
description=gr.Markdown(description),
fn=respond,
additional_inputs=[
gr.Textbox(value="You are Minitron an AI assistant created by Tonic-AI", label="System message"),
gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"),
gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
gr.Slider(minimum=0.1, maximum=1.0, value=0.95, step=0.05, label="Top-p (nucleus sampling)")
],
)
if __name__ == "__main__":
demo.launch() |