docs/source/en/model_doc/nemotron.md
This model was released on 2024-02-26 and added to Hugging Face Transformers on 2024-08-06.
Minitron is released under the NVIDIA Open Model License Agreement. The use of this model is governed by the NVIDIA AI Foundation Models Community License Agreement.
Nemotron-4 is a family of enterprise ready generative text models compatible with NVIDIA NeMo Framework.
NVIDIA NeMo is an end-to-end, cloud-native platform to build, customize, and deploy generative AI models anywhere. It includes training and inferencing frameworks, guardrailing toolkits, data curation tools, and pretrained models, offering enterprises an easy, cost-effective, and fast way to adopt generative AI. To get access to NeMo Framework, please sign up at this link.
Architecture Type: Transformer
Network Architecture: Transformer Decoder (auto-regressive language model).
Minitron is a family of small language models (SLMs) obtained by pruning NVIDIA's 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.
Deriving the Minitron 8B and 4B models from the base 15B model using our approach requires up to 40x fewer training tokens per model compared to training from scratch; this results in compute cost savings of 1.8x for training the full model family (15B, 8B, and 4B). Minitron models exhibit up to a 16% improvement in MMLU scores compared to training from scratch, perform comparably to other community models such as Mistral 7B, Gemma 7B and Llama-3 8B, and outperform state-of-the-art compression techniques from the literature. Please refer to our arXiv paper for more details.
Minitron models are for research and development only.
The following code provides an example of how to load the Minitron-4B model and use it to perform text generation.
from transformers import AutoModelForCausalLM, AutoTokenizer
# Load the tokenizer and model
model_path = 'nvidia/Minitron-4B-Base'
tokenizer = AutoTokenizer.from_pretrained(model_path)
model = AutoModelForCausalLM.from_pretrained(model_path, device_map="auto")
# Prepare the input text
prompt = 'Complete the paragraph: our solar system is'
inputs = tokenizer.encode(prompt, return_tensors='pt').to(model.device)
# Generate the output
outputs = model.generate(inputs, max_length=20)
# Decode and print the output
output_text = tokenizer.decode(outputs[0])
print(output_text)
5-shot performance. Language Understanding evaluated using Massive Multitask Language Understanding:
| Average |
|---|
| 58.6 |
Zero-shot performance. Evaluated using select datasets from the LM Evaluation Harness with additions:
| HellaSwag | Winogrande | GSM8K | ARC-C | XLSum |
|---|---|---|---|---|
| 75.0 | 74.0 | 24.1 | 50.9 | 29.5 |
Code generation performance. Evaluated using HumanEval:
| p@1, 0-Shot |
|---|
| 23.3 |
Please refer to our paper for the full set of results.
If you find our work helpful, please consider citing our paper:
@article{minitron2024,
title={Compact Language Models via Pruning and Knowledge Distillation},
author={Saurav Muralidharan and Sharath Turuvekere Sreenivas and Raviraj Joshi and Marcin Chochowski and Mostofa Patwary and Mohammad Shoeybi and Bryan Catanzaro and Jan Kautz and Pavlo Molchanov},
journal={arXiv preprint arXiv:2407.14679},
year={2024},
url={https://huggingface.co/papers/2407.14679},
}
[[autodoc]] NemotronConfig
[[autodoc]] NemotronModel - forward
[[autodoc]] NemotronForCausalLM - forward
[[autodoc]] NemotronForSequenceClassification - forward
[[autodoc]] NemotronForQuestionAnswering - forward
[[autodoc]] NemotronForTokenClassification - forward