File size: 6,618 Bytes
33e9db5 78e6344 33e9db5 |
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 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 |
---
language:
- en
- fr
- es
- pt
tags:
- falcon3
base_model: tiiuae/Falcon3-7B-Base
license: other
license_name: falcon-llm-license
license_link: https://falconllm.tii.ae/falcon-terms-and-conditions.html
---
<div align="center">
<img src="https://huggingface.co/datasets/tiiuae/documentation-images/resolve/main/general/falco3-logo.png" alt="drawing" width="500"/>
</div>
# Falcon3-7B-Instruct
**Falcon3** family of Open Foundation Models is a set of pretrained and instruct LLMs ranging from 1B to 10B.
This repository contains the **Falcon3-7B-Instruct**. It achieves state of art results (at release's time) on reasoning, language understanding, instruction following, code and mathematics tasks.
Falcon3-7B-Instruct supports 4 languages (english, french, spanish, portuguese) and a context length up to 32K.
## Model Details
- Architecture
- Transformer based causal decoder only architecture
- 28 decoder blocks
- Grouped query attention (GQA) for faster inference: 12 query heads and 4 key value heads
- Wider head dimension: 256
- High RoPE value to support long context understanding: 1000042
- Uses SwiGLU and RMSNorm
- 32K context length
- 131K vocab size
- Pretrained on 14 Teratokens of datasets comprising of web, code, STEM, high quality and mutlilingual data using 2048 H100 GPU chips
- Postrained on 1.2 million samples of STEM, conversations, code, safety and function call data
- Supports EN, FR, ES, PT
- Developed by [Technology Innovation Institute](https://www.tii.ae)
- License: TII Falcon-LLM License 2.0
- Model Release Date: December 2024
## Getting started
<details>
<summary> Click to expand </summary>
```python
from transformers import AutoTokenizer, AutoModelForCausalLM
from transformers import AutoModelForCausalLM, AutoTokenizer
model_name = "tiiuae/Falcon3-7B-Instruct"
model = AutoModelForCausalLM.from_pretrained(
model_name,
torch_dtype="auto",
device_map="auto"]
)
tokenizer = AutoTokenizer.from_pretrained(model_name)
prompt = "How many hours in one day?"
messages = [
{"role": "system", "content": "You are a helpful friendly assistant Falcon3 from TII, try to follow instructions as much as possible."},
{"role": "user", "content": prompt}
]
text = tokenizer.apply_chat_template(
messages,
tokenize=False,
add_generation_prompt=True
)
model_inputs = tokenizer([text], return_tensors="pt").to(model.device)
generated_ids = model.generate(
**model_inputs,
max_new_tokens=1024
)
generated_ids = [
output_ids[len(input_ids):] for input_ids, output_ids in zip(model_inputs.input_ids, generated_ids)
]
response = tokenizer.batch_decode(generated_ids, skip_special_tokens=True)[0]
print(response)
```
</details>
<br>
## Benchmarks
We report in the following table our internal pipeline benchmarks:
<table border="1" style="width: 100%; text-align: center; border-collapse: collapse;">
<colgroup>
<col style="width: 10%;">
<col style="width: 10%;">
<col style="width: 7%;">
<col style="width: 7%;">
<col style="background-color: rgba(80, 15, 213, 0.5); width: 7%;">
</colgroup>
<thead>
<tr>
<th>Category</th>
<th>Benchmark</th>
<th>Llama-3.1-8B-Instruct</th>
<th>Qwen2.5-7B-Instruct</th>
<th>Falcon3-7B-Instruct</th>
</tr>
</thead>
<tbody>
<tr>
<td rowspan="3">General</td>
<td>MMLU (5-shot)</td>
<td>55.9</td>
<td><b>72.4</b></td>
<td>68</td>
</tr>
<tr>
<td>MMLU-PRO (5-shot)</td>
<td>21.8</td>
<td>35.8</td>
<td><b>40.7</b></td>
</tr>
<tr>
<td>IFEval</td>
<td><b>78.8</b></td>
<td>74.7</td>
<td>76.5</td>
</tr>
<tr>
<td rowspan="3">Math</td>
<td>GSM8K (5-shot)</td>
<td>78.1</td>
<td>77.5</td>
<td><b>79.1</b></td>
</tr>
<tr>
<td>GSM8K (8-shot, COT)</td>
<td>79.8</td>
<td>72.7</td>
<td><b>80.9</b></td>
</tr>
<tr>
<td>MATH Lvl-5 (4-shot)</td>
<td>10.4</td>
<td>26</td>
<td><b>33.1</b></td>
</tr>
<tr>
<td rowspan="5">Reasoning</td>
<td>Arc Challenge (25-shot)</td>
<td>46.6</td>
<td>55.7</td>
<td><b>65.9</b></td>
</tr>
<tr>
<td>GPQA (0-shot)</td>
<td><b>33.6</b></td>
<td>31.9</td>
<td>32</td>
</tr>
<tr>
<td>GPQA (0-shot, COT)</td>
<td>9.6</td>
<td>13.8</td>
<td><b>22.3</b></td>
</tr>
<tr>
<td>MUSR (0-shot)</td>
<td>38.6</td>
<td>40.7</td>
<td><b>46.4</b></td>
</tr>
<tr>
<td>BBH (3-shot)</td>
<td>43.7</td>
<td><b>53.9</b></td>
<td>52.4</td>
</tr>
<tr>
<td rowspan="4">CommonSense Understanding</td>
<td>PIQA (0-shot)</td>
<td><b>78.9</b></td>
<td>73.7</td>
<td>78.8</td>
</tr>
<tr>
<td>SciQ (0-shot)</td>
<td>80.2</td>
<td>50.9</td>
<td><b>94.7</b></td>
</tr>
<tr>
<td>Winogrande (0-shot)</td>
<td>-</td>
<td>-</td>
<td>70.4</td>
</tr>
<tr>
<td>OpenbookQA (0-shot)</td>
<td><b>46.2</b></td>
<td>42.4</td>
<td>45.8</td>
</tr>
<tr>
<td rowspan="2">Instructions following</td>
<td>MT-Bench (avg)</td>
<td>7.86</td>
<td><b>8.54</b></td>
<td>8.36</td>
</tr>
<tr>
<td>Alapaca (WC)</td>
<td>26.57</td>
<td><b>31.5</b></td>
<td>26.13</td>
</tr>
<tr>
<td>Tool use</td>
<td>BFCL AST (avg)</td>
<td>90.6</td>
<td><b>91.4</b></td>
<td>72.3</td>
</tr>
</tbody>
</table>
## Technical Report
Coming soon....
## Citation
If Falcon3 family were helpful to your work, feel free to give us a cite.
```
@misc{Falcon3,
title = {The Falcon 3 family of Open Models},
author = {TII Team},
month = {December},
year = {2024}
}
```
|