File size: 34,583 Bytes
f17d263 |
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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 |
Quantization made by Richard Erkhov.
[Github](https://github.com/RichardErkhov)
[Discord](https://discord.gg/pvy7H8DZMG)
[Request more models](https://github.com/RichardErkhov/quant_request)
jais-adapted-7b-chat - GGUF
- Model creator: https://huggingface.co/inceptionai/
- Original model: https://huggingface.co/inceptionai/jais-adapted-7b-chat/
| Name | Quant method | Size |
| ---- | ---- | ---- |
| [jais-adapted-7b-chat.Q2_K.gguf](https://huggingface.co/RichardErkhov/inceptionai_-_jais-adapted-7b-chat-gguf/blob/main/jais-adapted-7b-chat.Q2_K.gguf) | Q2_K | 2.5GB |
| [jais-adapted-7b-chat.IQ3_XS.gguf](https://huggingface.co/RichardErkhov/inceptionai_-_jais-adapted-7b-chat-gguf/blob/main/jais-adapted-7b-chat.IQ3_XS.gguf) | IQ3_XS | 2.76GB |
| [jais-adapted-7b-chat.IQ3_S.gguf](https://huggingface.co/RichardErkhov/inceptionai_-_jais-adapted-7b-chat-gguf/blob/main/jais-adapted-7b-chat.IQ3_S.gguf) | IQ3_S | 2.9GB |
| [jais-adapted-7b-chat.Q3_K_S.gguf](https://huggingface.co/RichardErkhov/inceptionai_-_jais-adapted-7b-chat-gguf/blob/main/jais-adapted-7b-chat.Q3_K_S.gguf) | Q3_K_S | 2.9GB |
| [jais-adapted-7b-chat.IQ3_M.gguf](https://huggingface.co/RichardErkhov/inceptionai_-_jais-adapted-7b-chat-gguf/blob/main/jais-adapted-7b-chat.IQ3_M.gguf) | IQ3_M | 3.05GB |
| [jais-adapted-7b-chat.Q3_K.gguf](https://huggingface.co/RichardErkhov/inceptionai_-_jais-adapted-7b-chat-gguf/blob/main/jais-adapted-7b-chat.Q3_K.gguf) | Q3_K | 3.22GB |
| [jais-adapted-7b-chat.Q3_K_M.gguf](https://huggingface.co/RichardErkhov/inceptionai_-_jais-adapted-7b-chat-gguf/blob/main/jais-adapted-7b-chat.Q3_K_M.gguf) | Q3_K_M | 3.22GB |
| [jais-adapted-7b-chat.Q3_K_L.gguf](https://huggingface.co/RichardErkhov/inceptionai_-_jais-adapted-7b-chat-gguf/blob/main/jais-adapted-7b-chat.Q3_K_L.gguf) | Q3_K_L | 3.5GB |
| [jais-adapted-7b-chat.IQ4_XS.gguf](https://huggingface.co/RichardErkhov/inceptionai_-_jais-adapted-7b-chat-gguf/blob/main/jais-adapted-7b-chat.IQ4_XS.gguf) | IQ4_XS | 3.56GB |
| [jais-adapted-7b-chat.Q4_0.gguf](https://huggingface.co/RichardErkhov/inceptionai_-_jais-adapted-7b-chat-gguf/blob/main/jais-adapted-7b-chat.Q4_0.gguf) | Q4_0 | 3.73GB |
| [jais-adapted-7b-chat.IQ4_NL.gguf](https://huggingface.co/RichardErkhov/inceptionai_-_jais-adapted-7b-chat-gguf/blob/main/jais-adapted-7b-chat.IQ4_NL.gguf) | IQ4_NL | 3.75GB |
| [jais-adapted-7b-chat.Q4_K_S.gguf](https://huggingface.co/RichardErkhov/inceptionai_-_jais-adapted-7b-chat-gguf/blob/main/jais-adapted-7b-chat.Q4_K_S.gguf) | Q4_K_S | 3.76GB |
| [jais-adapted-7b-chat.Q4_K.gguf](https://huggingface.co/RichardErkhov/inceptionai_-_jais-adapted-7b-chat-gguf/blob/main/jais-adapted-7b-chat.Q4_K.gguf) | Q4_K | 3.97GB |
| [jais-adapted-7b-chat.Q4_K_M.gguf](https://huggingface.co/RichardErkhov/inceptionai_-_jais-adapted-7b-chat-gguf/blob/main/jais-adapted-7b-chat.Q4_K_M.gguf) | Q4_K_M | 3.97GB |
| [jais-adapted-7b-chat.Q4_1.gguf](https://huggingface.co/RichardErkhov/inceptionai_-_jais-adapted-7b-chat-gguf/blob/main/jais-adapted-7b-chat.Q4_1.gguf) | Q4_1 | 4.12GB |
| [jais-adapted-7b-chat.Q5_0.gguf](https://huggingface.co/RichardErkhov/inceptionai_-_jais-adapted-7b-chat-gguf/blob/main/jais-adapted-7b-chat.Q5_0.gguf) | Q5_0 | 4.52GB |
| [jais-adapted-7b-chat.Q5_K_S.gguf](https://huggingface.co/RichardErkhov/inceptionai_-_jais-adapted-7b-chat-gguf/blob/main/jais-adapted-7b-chat.Q5_K_S.gguf) | Q5_K_S | 4.52GB |
| [jais-adapted-7b-chat.Q5_K.gguf](https://huggingface.co/RichardErkhov/inceptionai_-_jais-adapted-7b-chat-gguf/blob/main/jais-adapted-7b-chat.Q5_K.gguf) | Q5_K | 4.64GB |
| [jais-adapted-7b-chat.Q5_K_M.gguf](https://huggingface.co/RichardErkhov/inceptionai_-_jais-adapted-7b-chat-gguf/blob/main/jais-adapted-7b-chat.Q5_K_M.gguf) | Q5_K_M | 4.64GB |
| [jais-adapted-7b-chat.Q5_1.gguf](https://huggingface.co/RichardErkhov/inceptionai_-_jais-adapted-7b-chat-gguf/blob/main/jais-adapted-7b-chat.Q5_1.gguf) | Q5_1 | 4.91GB |
| [jais-adapted-7b-chat.Q6_K.gguf](https://huggingface.co/RichardErkhov/inceptionai_-_jais-adapted-7b-chat-gguf/blob/main/jais-adapted-7b-chat.Q6_K.gguf) | Q6_K | 5.35GB |
| [jais-adapted-7b-chat.Q8_0.gguf](https://huggingface.co/RichardErkhov/inceptionai_-_jais-adapted-7b-chat-gguf/blob/main/jais-adapted-7b-chat.Q8_0.gguf) | Q8_0 | 6.93GB |
Original model description:
---
language:
- ar
- en
thumbnail: null
tags:
- Arabic
- English
- LLM
- Decoder
- causal-lm
- jais-family
license: apache-2.0
pipeline_tag: text-generation
---
# Jais Family Model Card
The Jais family of models is a comprehensive series of bilingual English-Arabic large language models (LLMs). These models are optimized to excel in Arabic while having strong English capabilities. We release two variants of foundation models that include:
- Models **pre-trained from scratch** (`jais-family-*`).
- Models **pre-trained adaptively from [Llama-2](https://arxiv.org/pdf/2307.09288)** (`jais-adapted-*`).
In this release, we introduce 20 models across 8 sizes, ranging from 590M to 70B parameters, trained on up to 1.6T tokens of Arabic, English, and code data. *All* pre-trained models in this series are instruction fine-tuned (`*-chat`) for dialog using a curated mix of Arabic and English instruction data.
We hope this extensive release will accelerate research in Arabic NLP, and enable numerous downstream applications for the Arabic speaking and bilingual community. The training and adaptation techniques we demonstrate successfully for Arabic models are extensible to other low and medium resource languages.
## Jais Family Details
- **Developed by:** Inception, Cerebras Systems.
- **Language(s):** (NLP): Arabic (MSA) and English.
- **Input:** Text only data.
- **Output:** Model generates text.
- **Model Sizes:** 590M, 1.3B, 2.7B, 6.7B, 7B, 13B, 30B, 70B.
- **Demo:** [Access the live demo here](https://arabic-gpt.ai/)
- **License:** Apache 2.0
| **Pre-trained Model** | **Fine-tuned Model** | **Size (Parameters)** | **Context length (Tokens)** |
|:---------------------|:--------|:-------|:-------|
| [jais-family-30b-16k](https://huggingface.co/inceptionai/jais-family-30b-16k) | [Jais-family-30b-16k-chat](https://huggingface.co/inceptionai/jais-family-30b-16k-chat) | 30B | 16,384 |
| [jais-family-30b-8k](https://huggingface.co/inceptionai/jais-family-30b-8k) | [Jais-family-30b-8k-chat](https://huggingface.co/inceptionai/jais-family-30b-8k-chat) | 30B | 8,192 |
| [jais-family-13b ](https://huggingface.co/inceptionai/jais-family-13b) | [Jais-family-13b-chat](https://huggingface.co/inceptionai/jais-family-13b-chat) | 13B | 2,048 |
| [jais-family-6p7b](https://huggingface.co/inceptionai/jais-family-6p7b) | [Jais-family-6p7b-chat](https://huggingface.co/inceptionai/jais-family-6p7b-chat) | 6.7B | 2,048 |
| [jais-family-2p7b](https://huggingface.co/inceptionai/jais-family-2p7b) | [Jais-family-2p7b-chat](https://huggingface.co/inceptionai/jais-family-2p7b-chat) | 2.7B | 2,048 |
| [jais-family-1p3b](https://huggingface.co/inceptionai/jais-family-1p3b) | [Jais-family-1p3b-chat](https://huggingface.co/inceptionai/jais-family-1p3b-chat) | 1.3B | 2,048 |
| [jais-family-590m](https://huggingface.co/inceptionai/jais-family-590m) | [Jais-family-590m-chat](https://huggingface.co/inceptionai/jais-family-590m-chat) | 590M | 2,048 |
| **Adapted pre-trained Model** | **Fine-tuned Model** | **Size (Parameters)** | **Context length (Tokens)** |
|:---------------------|:--------|:-------|:-------|
| [jais-adapted-70b](https://huggingface.co/inceptionai/jais-adapted-70b) | [Jais-adapted-70b-chat](https://huggingface.co/inceptionai/jais-adapted-70b-chat) | 70B | 4,096 |
| [jais-adapted-13b](https://huggingface.co/inceptionai/jais-adapted-13b) | [Jais-adapted-13b-chat](https://huggingface.co/inceptionai/jais-adapted-13b-chat) | 13B | 4,096 |
| [jais-adapted-7b](https://huggingface.co/inceptionai/jais-adapted-7b) | [Jais-adapted-7b-chat](https://huggingface.co/inceptionai/jais-adapted-7b-chat) | 7B | 4,096 |
### Model Architecture:
<a name="model-architecture"></a>
All models in this family are auto-regressive language models that use a transformer-based, decoder-only architecture (GPT-3).
Jais models (`jais-family-*`) are *trained from scratch*, incorporating the SwiGLU non-linear activation function and ALiBi position encoding. These architectural enhancements allow the models to extrapolate at long sequence lengths, leading to improved context handling and precision.
Jais adapted models (`jais-adapted-*`) are *built on top of Llama-2*, which employs RoPE position embedding and Grouped Query Attention. We introduce tokenizer expansion with Arabic data, which improves fertility and compute efficiency by over 3x. In particular, we add `32,000` new Arabic tokens from the Jais-30b vocabulary into the Llama-2 tokenizer.
To initialize these new Arabic token embeddings we first learn a linear projection from the embedding space of Jais-30b to Llama's embedding space, using the set of shared English tokens present in both vocabularies. Next, this learned projection is applied to transform the existing Jais-30b Arabic embeddings into the Llama-2 embedding space.
## Getting started
Below is sample code to use the model. Note that the model requires a custom model class, so users must enable `trust_remote_code=True` while loading the model.
```python
# -*- coding: utf-8 -*-
import torch
from transformers import AutoTokenizer, AutoModelForCausalLM
model_path = "inceptionai/jais-adapted-7b-chat"
prompt_eng = "### Instruction:Your name is 'Jais', and you are named after Jebel Jais, the highest mountain in UAE. You were made by 'Inception' in the UAE. You are a helpful, respectful, and honest assistant. Always answer as helpfully as possible, while being safe. Complete the conversation between [|Human|] and [|AI|]:\n### Input: [|Human|] {Question}\n[|AI|]\n### Response :"
prompt_ar = "### Instruction:اسمك \"جيس\" وسميت على اسم جبل جيس اعلى جبل في الامارات. تم بنائك بواسطة Inception في الإمارات. أنت مساعد مفيد ومحترم وصادق. أجب دائمًا بأكبر قدر ممكن من المساعدة، مع الحفاظ على البقاء أمناً. أكمل المحادثة بين [|Human|] و[|AI|] :\n### Input:[|Human|] {Question}\n[|AI|]\n### Response :"
device = "cuda" if torch.cuda.is_available() else "cpu"
tokenizer = AutoTokenizer.from_pretrained(model_path)
model = AutoModelForCausalLM.from_pretrained(model_path, device_map="auto", trust_remote_code=True)
if tokenizer.pad_token is None:
tokenizer.pad_token = tokenizer.eos_token
def get_response(text, tokenizer=tokenizer, model=model):
tokenized = tokenizer(text, return_tensors="pt")
input_ids, attention_mask = tokenized['input_ids'].to(device), tokenized['attention_mask'].to(device)
input_len = input_ids.shape[-1]
generate_ids = model.generate(
input_ids,
attention_mask=attention_mask,
top_p=0.9,
temperature=0.3,
max_length=2048,
min_length=input_len + 4,
repetition_penalty=1.2,
do_sample=True,
pad_token_id=tokenizer.pad_token_id
)
response = tokenizer.batch_decode(
generate_ids, skip_special_tokens=True, clean_up_tokenization_spaces=True
)[0]
response = response.split("### Response :")[-1].lstrip()
return response
ques = "ما هي عاصمة الامارات؟"
text = prompt_ar.format_map({'Question': ques})
print(get_response(text))
ques = "What is the capital of UAE?"
text = prompt_eng.format_map({'Question': ques})
print(get_response(text))
```
## Training Details
### Pretraining Data
The Jais family of models are trained on up to 1.6 Trillion tokens of diverse English, Arabic and Code data. The data consists of the following sources:
- **Web:** We used publicly available web pages, wikipedia articles, news articles, and social network content in both Arabic and English.
- **Code:** To enhance the reasoning capability of our model, we include Code data in various programming languages.
- **Books:** We used a selection of publicly available Arabic and English books data, which improves long-range context modelling and coherent storytelling.
- **Scientific:** A subset of ArXiv papers were included to improve reasoning and long context abilities.
- **Synthetic:** We augment the volume of Arabic data by translating English to Arabic using an in-house machine translation system. We restrict this to high quality English resources such as English Wikipedia and English books.
We extensively preprocess and deduplicate the training data. For Arabic, we used a custom preprocessing pipeline to filter for data with high linguistic quality. More information on this pipeline can be found in the [Jais paper](https://arxiv.org/abs/2308.16149).
- **Jais pre-trained** (`jais-family-*`): Following our previous experimentation with language alignment mixing in [Jais](https://arxiv.org/abs/2308.16149), we used a ratio of 1:2:0.4 of Arabic:English:Code data. This recipe for <u>from scratch pre-training</u> addresses Arabic data scarcity while improving performance in both languages.
- **Jais adapted pre-trained** (`jais-adapted-*`): For the <u>adapted pre-training of Llama-2</u>, we utilized a larger Arabic dataset of ~334B Arabic tokens mixed with English and Code data. We vary the mixing ratio, at different model sizes, to introduce strong Arabic capabilities while maintaining performance in English.
| **Pre-trained model** | **English data (tokens)** | **Arabic data (tokens)** | **Code data (tokens)** | **Total data (tokens)** |
|-------------------------|---------------------------|--------------------------|------------------------|------------------------|
| [jais-family-30b-16k](https://huggingface.co/inceptionai/jais-family-30b-16k) | 980B | 490B | 196B | 1666B |
| [jais-family-30b-8k](https://huggingface.co/inceptionai/jais-family-30b-8k) | 882B | 441B | 177B | 1500B |
| [jais-family-13b ](https://huggingface.co/inceptionai/jais-family-13b) | 283B | 141B | 56B | 480B |
| [jais-family-6p7b](https://huggingface.co/inceptionai/jais-family-6p7b) | 283B | 141B | 56B | 480B |
| [jais-family-2p7b](https://huggingface.co/inceptionai/jais-family-2p7b) | 283B | 141B | 56B | 480B |
| [jais-family-1p3b](https://huggingface.co/inceptionai/jais-family-1p3b) | 283B | 141B | 56B | 480B |
| [jais-family-590m](https://huggingface.co/inceptionai/jais-family-590m) | 283B | 141B | 56B | 480B |
| [jais-adapted-70b](https://huggingface.co/inceptionai/jais-adapted-70b) | 33B | 334B | 4B | 371B |
| [jais-adapted-13b](https://huggingface.co/inceptionai/jais-adapted-13b) | 127B | 140B | 13B | 280B |
| [jais-adapted-7b](https://huggingface.co/inceptionai/jais-adapted-7b) | 18B | 19B | 2B | 39B |
### Finetuning data
<!-- This should link to a Data Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
All chat models in the Jais family are fine-tuned using Arabic and English prompt-response pairs in both single-turn and multi-turn settings. Data sources include open-source fine-tuning datasets filtered for topic and style diversity. Additionally, internally curated human data is incorporated to enhance cultural adaptation. This data is supplemented with content generated using synthetic methods including machine translation, distillation, and model self-chat. Overall, our updated instruction-tuning dataset comprises ~10M and ~4M prompt-response pairs in English and Arabic respectively.
### Training Procedure
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
During the pre-training of (`jais-family-*`) models, documents are packed into sequences separated by EOS tokens, and the model is trained autoregressively, applying the loss to all tokens. For jais-30b models, the context length is progressively expanded from 2k to 8K to 16K by incorporating curated long-context documents in training. This progressive expansion leverages faster initial training at shorter context lengths, while gradually extending support for larger context lengths towards the end of the training process.
During the adapted pre-training of the (`jais-adapted-*`) models, we first initialize the new tokenizer and Arabic embeddings as described in [Model Architecture](#model-architecture). In training, we implemented a two-stage approach to overcome observed higher norms of the new Arabic embeddings. In the first stage, the backbone of the model is frozen, and the embeddings are trained using approximately 15 billion tokens from a bilingual corpus of English and Arabic. In the second stage, the backbone is unfrozen, and continuous pretraining is conducted with all parameters.
During instruction tuning, each training example consists of a single-turn or multi-turn prompt and it's response. Instead of one example per sequence, examples are packed together while the loss is masked on the prompt tokens. This approach speeds up training by allowing more examples to be processed per batch.
### Training Hyperparameters:
#### Jais-adapted-7b-chat
| Hyperparameter | Value |
|----------------|-------------------------------------------|
| Precision | fp32 |
| Optimizer | AdamW |
| Learning rate | 0 to 2.0e-05(<=380 warmup steps)<br>2.0e-05 to 2.0e-06(>380 and <=13175 steps, Cosine Decay) |
| Weight decay | 0.1 |
| Batch size | 264|
| Context Length | 4096|
| Steps | 13175 |
### Compute Infrastructure
The training process was performed on the Condor Galaxy (CG) supercomputer platform. A CG contains 64 Cerebras CS-2 Wafer-Scale Engines (WSE-2) with 40 GB of SRAM, and achieves a total of 960 PetaFLOP/s.
## Evaluation
<!-- This section describes the evaluation protocols and provides the results. -->
We conducted a comprehensive evaluation of Jais models focusing on both English and Arabic, using LM-harness in a zero-shot setting. The evaluation criteria spanned various dimensions, including:
- **Knowledge:** How well the model answers factual questions.
- **Reasoning:** The model's ability to answer questions requiring reasoning.
- **Misinformation/Bias:** Assessment of the model's susceptibility to generating false or misleading information, and its neutrality.
### Arabic evaluation results:
<style>
.table-container {
overflow-x: auto;
white-space: nowrap;
}
</style>
<div class="table-container">
| **Models** | Avg | ArabicMMLU*| MMLU | EXAMS*| LitQA*| agqa | agrc | Hellaswag | PIQA | BoolQA | Situated QA | ARC-C | OpenBookQA | TruthfulQA | CrowS-Pairs |
|--------------------------|-------|------------|-------|-------|-------|------|------|------------|------|--------|-------------|-------|------------|------------|-------------|
| jais-family-30b-16k | 49.2 | 44.0 | 33.4 | 40.9 | 60 | 47.8 | 49.3 | 60.9 | 68.6 | 70.3 | 41.6 | 38.7 | 31.8 | 45.2 | 57 |
| jais-family-30b-8k | 49.7 | 46.0 | 34 | 42 | 60.6 | 47.6 | 50.4 | 60.4 | 69 | 67.7 | 42.2 | 39.2 | 33.8 | 45.1 | 57.3 |
| jais-family-13b | 46.1 | 34.0 | 30.3 | 42.7 | 58.3 | 40.5 | 45.5 | 57.3 | 68.1 | 63.1 | 41.6 | 35.3 | 31.4 | 41 | 56.1 |
| jais-family-6p7b | 44.6 | 32.2 | 29.9 | 39 | 50.3 | 39.2 | 44.1 | 54.3 | 66.8 | 66.5 | 40.9 | 33.5 | 30.4 | 41.2 | 55.4 |
| jais-family-2p7b | 41.0 | 29.5 | 28.5 | 36.1 | 45.7 | 32.4 | 40.8 | 44.2 | 62.5 | 62.2 | 39.2 | 27.4 | 28.2 | 43.6 | 53.6 |
| jais-family-1p3b | 40.8 | 28.9 | 28.5 | 34.2 | 45.7 | 32.4 | 40.8 | 44.2 | 62.5 | 62.2 | 39.2 | 27.4 | 28.2 | 43.6 | 53.6 |
| jais-family-590m | 39.7 | 31.2 | 27 | 33.1 | 41.7 | 33.8 | 38.8 | 38.2 | 60.7 | 62.2 | 37.9 | 25.5 | 27.4 | 44.7 | 53.3 |
| jais-family-30b-16k-chat | 51.6 | 59.9 | 34.6 | 40.2 | 58.9 | 46.8 | 54.7 | 56.2 | 64.4 | 76.7 | 55.9 | 40.8 | 30.8 | 49.5 | 52.9 |
| jais-family-30b-8k-chat | 51.4 | 61.2 | 34.2 | 40.2 | 54.3 | 47.3 | 53.6 | 60 | 63.4 | 76.8 | 54.7 | 39.5 | 30 | 50.7 | 54.3 |
| jais-family-13b-chat | 50.3 | 58.2 | 33.9 | 42.9 | 53.1 | 46.8 | 51.7 | 59.3 | 65.4 | 75.2 | 51.2 | 38.4 | 29.8 | 44.8 | 53.8 |
| jais-family-6p7b-chat | 48.7 | 55.7 | 32.8 | 37.7 | 49.7 | 40.5 | 50.1 | 56.2 | 62.9 | 79.4 | 52 | 38 | 30.4 | 44.7 | 52 |
| jais-family-2p7b-chat | 45.6 | 50.0 | 31.5 | 35.9 | 41.1 | 37.3 | 42.1 | 48.6 | 63.7 | 74.4 | 50.9 | 35.3 | 31.2 | 44.5 | 51.3 |
| jais-family-1p3b-chat | 42.7 | 42.2 | 30.1 | 33.6 | 40.6 | 34.1 | 41.2 | 43 | 63.6 | 69.3 | 44.9 | 31.6 | 28 | 45.6 | 50.4 |
| jais-family-590m-chat | 37.8 | 39.1 | 28 |29.5 | 33.1 | 30.8 | 36.4 | 30.3 | 57.8 | 57.2 | 40.5 | 25.9 | 26.8 | 44.5 | 49.3 |
| **Adapted Models** | Avg | ArabicMMLU*| MMLU | EXAMS*| LitQA*| agqa | agrc | Hellaswag | PIQA | BoolQA | Situated QA | ARC-C | OpenBookQA | TruthfulQA | CrowS-Pairs |
|--------------------------|-------|------------|-------|-------|-------|------|------|------------|------|--------|-------------|-------|------------|------------|-------------|
| jais-adapted-70b | 51.5 | 55.9 | 36.8 | 42.3 | 58.3 | 48.6 | 54 | 61.5 | 68.4 | 68.4 | 42.1 | 42.6 | 33 | 50.2 | 58.3 |
| jais-adapted-13b | 46.6 | 44.7 | 30.6 | 37.7 | 54.3 | 43.8 | 48.3 | 54.9 | 67.1 | 64.5 | 40.6 | 36.1 | 32 | 43.6 | 54.00 |
| jais-adapted-7b | 42.0 | 35.9 | 28.9 | 36.7 | 46.3 | 34.1 | 40.3 | 45 | 61.3 | 63.8 | 38.1 | 29.7 | 30.2 | 44.3 | 53.6 |
| jais-adapted-70b-chat | 52.9 | 66.8 | 34.6 | 42.5 | 62.9 | 36.8 | 48.6 | 64.5 | 69.7 | 82.8 | 49.3 | 44.2 | 32.2 | 53.3 | 52.4 |
| jais-adapted-13b-chat | 50.3 | 59.0 | 31.7 | 37.5 | 56.6 | 41.9 | 51.7 | 58.8 | 67.1 | 78.2 | 45.9 | 41 | 34.2 | 48.3 | 52.1 |
| jais-adapted-7b-chat | 46.1 | 51.3 | 30 | 37 | 48 | 36.8 | 48.6 | 51.1 | 62.9 | 72.4 | 41.3 | 34.6 | 30.4 | 48.6 | 51.8 |
</div>
Arabic benchmarks are translated using an in-house MT model and reviewed by Arabic linguists. Benchmarks labeled with an asterisk (*) are natively Arabic; for further details, see the [Jais paper](https://arxiv.org/abs/2308.16149). Additionally, we include [ArabicMMLU](https://arxiv.org/abs/2402.12840), a native Arabic benchmark based on regional knowledge.
### English evaluation results:
<div class="table-container">
| **Models** | Avg | MMLU | RACE | Hellaswag | PIQA | BoolQA | SIQA | ARC-Challenge | OpenBookQA | Winogrande | TruthfulQA | CrowS-Pairs |
|--------------------------|----------|------|------|-----------|------|--------|------|---------------|------------|------------|----------------|-------------|
| jais-family-30b-16k | 59.3 | 42.2 | 40.5 | 79.7 | 80.6 | 78.7 | 48.8 | 50.3 | 44.2 | 71.6 | 43.5 | 72.6 |
| jais-family-30b-8k | 58.8 | 42.3 | 40.3 | 79.1 | 80.5 | 80.9 | 49.3 | 48.4 | 43.2 | 70.6 | 40.3 | 72.3 |
| jais-family-13b | 54.6 | 32.3 | 39 | 72 | 77.4 | 73.9 | 47.9 | 43.2 | 40 | 67.1 | 36.1 | 71.7 |
| jais-family-6p7b | 53.1 | 32 | 38 | 69.3 | 76 | 71.7 | 47.1 | 40.3 | 37.4 | 65.1 | 34.4 | 72.5 |
| jais-family-2p7b | 51 | 29.4 | 38 | 62.7 | 74.1 | 67.4 | 45.6 | 35.1 | 35.6 | 62.9 | 40.1 | 70.2 |
| jais-family-1p3b | 48.7 | 28.2 | 35.4 | 55.4 | 72 | 62.7 | 44.9 | 30.7 | 36.2 | 60.9 | 40.4 | 69 |
| jais-family-590m | 45.2 | 27.8 | 32.9 | 46.1 | 68.1 | 60.4 | 43.2 | 25.6 | 30.8 | 55.8 | 40.9 | 65.3 |
| jais-family-30b-16k-chat | 58.8 | 42 | 41.1 | 76.2 | 73.3 | 84.6 | 60.3 | 48.4 | 40.8 | 68.2 | 44.8 | 67 |
| jais-family-30b-8k-chat | 60.3 | 40.6 | 47.1 | 78.9 | 72.7 | 90.6 | 60 | 50.1 | 43.2 | 70.6 | 44.9 | 64.2 |
| jais-family-13b-chat | 57.5 | 36.6 | 42.6 | 75 | 75.8 | 87.6 | 54.4 | 47.9 | 42 | 65 | 40.6 | 64.5 |
| jais-family-6p7b-chat | 56 | 36.6 | 41.3 | 72 | 74 | 86.9 | 55.4 | 44.6 | 40 | 62.4 | 41 | 62.2 |
| jais-family-2p7b-chat | 52.8 | 32.7 | 40.4 | 62.2 | 71 | 84.1 | 54 | 37.2 | 36.8 | 61.4 | 40.9 | 59.8 |
| jais-family-1p3b-chat | 49.3 | 31.9 | 37.4 | 54.5 | 70.2 | 77.8 | 49.8 | 34.4 | 35.6 | 52.7 | 37.2 | 60.8 |
| jais-family-590m-chat | 42.6 | 27.9 | 33.4 | 33.1 | 63.7 | 60.1 | 45.3 | 26.7 | 25.8 | 50.5 | 44.5 | 57.7 |
</div>
<div class="table-container">
|**Adapted Models**| Avg | MMLU | RACE | Hellaswag | PIQA | BoolQA | SIQA | ARC-Challenge | OpenBookQA | Winogrande | TruthfulQA | CrowS-Pairs |
|--------------------------|----------|------|------|-----------|------|--------|------|---------------|------------|------------|----------------|-------------|
| jais-adapted-70b | 60.1 | 40.4 | 38.5 | 81.2 | 81.1 | 81.2 | 48.1 | 50.4 | 45 | 75.8 | 45.7 | 74 |
| jais-adapted-13b | 56 | 33.8 | 39.5 | 76.5 | 78.6 | 77.8 | 44.6 | 45.9 | 44.4 | 71.4 | 34.6 | 69 |
| jais-adapted-7b | 55.7 | 32.2 | 39.8 | 75.3 | 78.8 | 75.7 | 45.2 | 42.8 | 43 | 68 | 38.3 | 73.1 |
| jais-adapted-70b-chat | 61.4 | 38.7 | 42.9 | 82.7 | 81.2 | 89.6 | 52.9 | 54.9 | 44.4 | 75.7 | 44 | 68.8 |
| jais-adapted-13b-chat | 58.5 | 34.9 | 42.4 | 79.6 | 79.7 | 88.2 | 50.5 | 48.5 | 42.4 | 70.3 | 42.2 | 65.1 |
| jais-adapted-7b-chat | 58.5 | 33.8 | 43.9 | 77.8 | 79.4 | 87.1 | 47.3 | 46.9 | 43.4 | 69.9 | 42 | 72.4 |
</div>
### GPT-4 evaluation
In addition to the LM-Harness evaluation, we conducted an open-ended generation evaluation using GPT-4-as-a-judge. We measured pairwise win-rates of model responses in both Arabic and English on a fixed set of 80 prompts from the Vicuna test set.
English prompts were translated to Arabic by our in-house linguists.
In the following, we compare the models in this release of the jais family against previously released versions:
<p align="center">
<img src="https://huggingface.co/inceptionai/jais-family-30b-16k-chat/resolve/main/jais.png" alt="Jais-adapted GPT-4">
</p>
<p align="center">
<em>GPT-4-as-a-judge evaluation of Jais in Arabic and English. Jais family models are significantly better than previous Jais at generations in both languages. </em>
</p>
<p align="center">
<img src="https://huggingface.co/inceptionai/jais-family-30b-16k-chat/resolve/main/jais-adapted.png" alt="Jais-adapted GPT-4">
</p>
<p align="center">
<em>GPT-4-as-a-judge evaluation of adapted Jais in Arabic and English. The generation quality of Arabic is significantly enhanced, while achieving improvement in English when compared to Llama-2 instruct. </em>
</p>
Besides pairwise comparison, we also perform MT-bench style single-answer grading on a scale of 1 to 10.
<p align="center">
<img src="https://huggingface.co/inceptionai/jais-family-30b-16k-chat/resolve/main/mt_bench.png" alt="MT-bench">
</p>
<p align="center">
<em>MT-bench style single-answer grading evaluation of Jais and adapted Jais in Arabic and English. Comparisons are made between select corresponding models from earlier releases. The quality ratings of responses are generally improved, with significant enhancements in Arabic.</em>
</p>
## Intended use
We release the Jais family of models under a full open-source license. We welcome all feedback and opportunities to collaborate. Spanning sizes from 590M to 70B parameters, this suite of bilingual models accommodates a wide range of use cases. Some potential downstream applications include:
- **Research**: The Jais family serves Arabic researchers and NLP practitioners, offering both compute-efficient and advanced model sizes
- Natural language understanding and generation tasks.
- Mechanistic interpretability analyses on cultural alignment in bilingual pre-trained and adapted pre-trained models.
- Quantitative studies of Arabic cultural and linguistic phenomena.
- **Commercial Use**: Jais 30B and 70B chat models are well-suited for direct use in chat applications with appropriate prompting or for further fine-tuning on specific tasks.
- Development of chat assistants for Arabic-speaking users.
- Sentiment analysis to gain insights into local markets and customer trends.
- Summarization of bilingual Arabic-English documents.
Audiences that we hope will benefit from our model:
- **Academics**: For those researching Arabic Natural Language Processing.
- **Businesses**: Companies targeting Arabic-speaking audiences.
- **Developers**: Those integrating Arabic language capabilities in applications.
### Out-of-Scope Use
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
While the Jais family of models are powerful Arabic and English bilingual models, it's essential to understand their limitations
and the potential of misuse. It is prohibited to use the model in any manner that violates applicable laws or regulations.
The following are some example scenarios where the model should not be used.
- **Malicious Use**: The model should not be used to generate harmful, misleading, or inappropriate content. Thisincludes but is not limited to:
- Generating or promoting hate speech, violence, or discrimination.
- Spreading misinformation or fake news.
- Engaging in or promoting illegal activities.
- **Sensitive Information**: The model should not be used to handle or generate personal, confidential, or sensitive information.
- **Generalization Across All Languages**: Jais family of models are bilingual and optimized for Arabic and English. They should not be presumed to have equal proficiency in other languages or dialects.
- **High-Stakes Decisions**: The model should not be used to make high-stakes decisions without human oversight. This includes medical, legal, financial, or safety-critical decisions.
## Bias, Risks, and Limitations
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
The Jais family is trained on publicly available data which was in part curated by Inception. We have employed different techniques to reduce bias in the model. While efforts have been made to minimize biases, it is likely that the model, as with all LLM models, will exhibit some bias.
The fine-tuned variants are trained as an AI assistant for Arabic and English speakers. Chat models are limited to produce responses for queries in these two languages and may not produce appropriate responses to other language queries.
By using Jais, you acknowledge and accept that, as with any large language model, it may generate incorrect, misleading, and/or offensive information or content. The information is not intended as advice and should not be relied upon in any way, nor are we responsible for any of the content or consequences resulting from its use.
#### Summary
We release the Jais family of Arabic and English bilingual models. The wide range of pre-trained model sizes, the recipe for adapting English-centric models to Arabic, and the fine-tuning of all sizes unlocks numerous use cases commercially and academically in the Arabic setting.
Through this release, we aim to make LLMs more accessible to Arabic NLP researchers and companies, offering native Arabic models that provide better cultural understanding than English centric ones. The strategies we employ for pre-training, fine-tuning and adaptation to Arabic are extensible to other low and medium resource languages, paving the way for language-focused and accessible models that cater to local contexts.
#### Citation info
```bibtex
@misc{sengupta2023jais,
title={Jais and Jais-chat: Arabic-Centric Foundation and Instruction-Tuned Open Generative Large Language Models},
author={Neha Sengupta, Sunil Kumar Sahu, Bokang Jia, Satheesh Katipomu, Haonan Li, Fajri Koto, William Marshall, Gurpreet Gosal, Cynthia Liu, Zhiming Chen, Osama Mohammed Afzal, Samta Kamboj, Onkar Pandit, Rahul Pal, Lalit Pradhan, Zain Muhammad Mujahid, Massa Baali, Xudong Han, Sondos Mahmoud Bsharat, Alham Fikri Aji, Zhiqiang Shen, Zhengzhong Liu, Natalia Vassilieva, Joel Hestness, Andy Hock, Andrew Feldman, Jonathan Lee, Andrew Jackson, Hector Xuguang Ren, Preslav Nakov, Timothy Baldwin and Eric Xing},
year={2023},
eprint={2308.16149},
archivePrefix={arXiv},
primaryClass={cs.CL}
}
@article{jaisfamilymodelcard,
title={Jais Family Model Card},
author={Inception},
year={2024},
url = {https://huggingface.co/inceptionai/jais-family-30b-16k-chat/blob/main/README.md}
}
```
|