SmolLM2
Table of Contents
Model Summary
SmolLM2 is a family of compact language models available in three size: 135M, 360M, and 1.7B parameters. They are capable of solving a wide range of tasks while being lightweight enough to run on-device.
SmolLM2 demonstrates significant advances over its predecessor SmolLM1, particularly in instruction following, knowledge, reasoning. The 135M model was trained on 2 trillion tokens using a diverse dataset combination: FineWeb-Edu, DCLM, The Stack, along with new filtered datasets we curated and will release soon. We developed the instruct version through supervised fine-tuning (SFT) using a combination of public datasets and our own curated datasets. We then applied Direct Preference Optimization (DPO) using UltraFeedback.
The instruct model additionally supports tasks such as text rewriting, summarization and function calling thanks to datasets developed by Argilla such as Synth-APIGen-v0.1.
How to use
pip install transformers
Running the model on CPU/GPU/multi GPU
- Using full precision
# pip install transformers
from transformers import AutoModelForCausalLM, AutoTokenizer
checkpoint = "HuggingFaceTB/SmolLM2-135M"
device = "cuda" # for GPU usage or "cpu" for CPU usage
tokenizer = AutoTokenizer.from_pretrained(checkpoint)
# for multiple GPUs install accelerate and do `model = AutoModelForCausalLM.from_pretrained(checkpoint, device_map="auto")`
model = AutoModelForCausalLM.from_pretrained(checkpoint).to(device)
inputs = tokenizer.encode("Gravity is", return_tensors="pt").to(device)
outputs = model.generate(inputs)
print(tokenizer.decode(outputs[0]))
- Using
torch.bfloat16
# pip install accelerate
import torch
from transformers import AutoTokenizer, AutoModelForCausalLM
checkpoint = "HuggingFaceTB/SmolLM2-135M"
tokenizer = AutoTokenizer.from_pretrained(checkpoint)
# for fp16 use `torch_dtype=torch.float16` instead
model = AutoModelForCausalLM.from_pretrained(checkpoint, device_map="auto", torch_dtype=torch.bfloat16)
inputs = tokenizer.encode("Gravity is", return_tensors="pt").to("cuda")
outputs = model.generate(inputs)
print(tokenizer.decode(outputs[0]))
>>> print(f"Memory footprint: {model.get_memory_footprint() / 1e6:.2f} MB")
Memory footprint: 723.56 MB
Evaluation
In this section, we report the evaluation results of SmolLM2. All evaluations are zero-shot unless stated otherwise, and we use lighteval to run them.
Base pre-trained model
| Metrics | SmolLM2-135M-8k | SmolLM-135M |
|---|---|---|
| HellaSwag | 42.1 | 41.2 |
| ARC (Average) | 43.9 | 42.4 |
| PIQA | 68.4 | 68.4 |
| MMLU (cloze) | 31.5 | 30.2 |
| CommonsenseQA | 33.9 | 32.7 |
| TriviaQA | 4.1 | 4.3 |
| Winogrande | 51.3 | 51.3 |
| OpenBookQA | 34.6 | 34.0 |
| GSM8K (5-shot) | 1.4 | 1.0 |
Instruction model
| Metric | SmolLM2-135M-Instruct | SmolLM-135M-Instruct |
|---|---|---|
| IFEval (Average prompt/inst) | 29.9 | 17.2 |
| MT-Bench | 1.98 | 1.68 |
| HellaSwag | 40.9 | 38.9 |
| ARC (Average) | 37.3 | 33.9 |
| PIQA | 66.3 | 64.0 |
| MMLU (cloze) | 29.3 | 28.3 |
| BBH (3-shot) | 28.2 | 25.2 |
| GSM8K (5-shot) | 1.4 | 1.4 |
Limitations
SmolLM2 models primarily understand and generate content in English. They can produce text on a variety of topics, but the generated content may not always be factually accurate, logically consistent, or free from biases present in the training data. These models should be used as assistive tools rather than definitive sources of information. Users should always verify important information and critically evaluate any generated content.
Training
Model
- Architecture: Transformer decoder
- Pretraining tokens: 2T
- Precision: bfloat16
Hardware
- GPUs: 64 H100
Software
- Training Framework: nanotron
License
Citation
@misc{allal2024SmolLM2,
title={SmolLM2 - with great data, comes great performance},
author={Loubna Ben Allal and Anton Lozhkov and Elie Bakouch and Gabriel Mart铆n Bl谩zquez and Lewis Tunstall and Agust铆n Piqueres and Andres Marafioti and Cyril Zakka and Leandro von Werra and Thomas Wolf},
year={2024},
}
- Downloads last month
- 1,053