language:
- en
- fr
- es
- pt
tags:
- falcon3
license: other
license_name: falcon-llm-license
license_link: https://falconllm.tii.ae/falcon-terms-and-conditions.html
Falcon3-3B-Base
Falcon3 family of Open Foundation Models is a set of pretrained and instruct LLMs ranging from 1B to 10B.
This repository contains the Falcon3-3B-Base. It achieves strong results on reasoning, language understanding, instruction following, code and mathematics tasks. Falcon3-3B-Base supports 4 languages (english, french, spanish, portuguese) and a context length up to 8K. Falcon3-3B-Base pruned (depth + width) from Falcon3-7B-Base, was effeciently trained on only 100 GT using a knowledge distillation objective.
⚠️ This is a raw, pretrained model, which should be further finetuned using SFT, RLHF, continued pretraining, etc. for most use cases.
Model Details
- Architecture
- Transformer based causal decoder only architecture
- 22 decoder blocks
- Grouped query attention (GQA) for faster inference: 12 query heads and 4 KV heads
- Wider head dimension: 256
- High RoPE value to support long context understanding: 1000042
- Uses SwiGLu and RMSNorm
- 8K context length
- 131K vocab size
- Pruned and Healed from Falcon3-7B-Base on only 100 Gigatokens of datasets comprising of web, code, STEM, high quality and mutlilingual data using 2048 H100 GPU chips
- Supports EN, FR, ES, PT
- Developed by Technology Innovation Institute
- License: TII Falcon-LLM License 2.0
- Model Release Date: December 2024
Getting started
Click to expand
import torch
from transformers import pipeline
pipe = pipeline(
"text-generation",
model="tiiuae/Falcon3-3B-Base",
torch_dtype=torch.bfloat16,
device_map="auto"
)
response = pipe("Question: How many hours in one day? Answer: ")
print(response[0]['generated_text'])
Benchmarks
We report in the following table our internal pipeline benchmarks:
| Category | Benchmark | Llama3.2-3B | Qwen2.5-3B | Minitron-4B | Falcon3-3B-Base |
|---|---|---|---|---|---|
| General | MMLU (5-shot) | 56.1 | 65.6 | 58.6 | 55.5 |
| MMLU-PRO (5-shot) | 24.9 | 31.99 | 26.21 | 28.77 | |
| IFEval | 12.83 | 27.0 | 22.81 | 27.67 | |
| Math | GSM8K (5-shot) | 26.68 | 68.99 | 25.7 | 63.91 |
| MATH Lvl-5 (4-shot) | 1.39 | 8.43 | 1.73 | 9.38 | |
| Reasoning | Arc Challenge (25-shot) | 50.76 | 55.54 | 50.34 | 54.86 |
| GPQA (0-shot) | 27.49 | 27.53 | 38.6 | 31.15 | |
| MUSR (0-shot) | 35.24 | 43.03 | 42.13 | 37.5 | |
| BBH (3-shot) | 38.59 | 46.12 | 40.85 | 44.23 | |
| CommonSense Understanding | PIQA (0-shot) | 77.42 | 78.89 | 78.29 | 75.62 |
| SciQ (0-shot) | 92.7 | 95.6 | 96.1 | 93.1 | |
| Winogrande (0-shot) | 69.69 | 68.82 | 68.35 | 64.64 | |
| OpenbookQA (0-shot) | 43.2 | 42.2 | 43.0 | 39.4 |
Useful links
- View our release blogpost.
- Feel free to join our discord server if you have any questions or to interact with our researchers and developers.
Technical Report
Coming soon....
Citation
If the Falcon3 family of models were helpful to your work, feel free to give us a cite.
@misc{Falcon3,
title = {The Falcon 3 Family of Open Models},
author = {Falcon-LLM Team},
month = {December},
year = {2024}
}