File size: 11,379 Bytes
bb11a33
07c92ca
 
 
2b079b2
 
 
 
 
0695f87
07c92ca
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
bb11a33
068743a
7f0e104
cd8ae35
382cef1
 
7f0e104
 
068743a
 
f74b172
7698e3e
068743a
 
00d886f
068743a
00d886f
068743a
 
00d886f
f74b172
068743a
dcfd917
4e4bf51
00d886f
d9aae8f
60adef1
 
 
2b079b2
 
 
 
 
 
6625d6d
2b079b2
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
9442df0
2b079b2
 
 
 
068743a
0dfb32e
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
ea64f23
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
068743a
 
7d698b4
068743a
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
00ae028
 
068743a
 
 
 
 
d3167df
 
 
 
 
ee0b265
 
a8b6a90
 
 
12acc01
53b019b
5c6538b
4e4bf51
 
 
 
dcfd917
4e4bf51
 
 
dcfd917
4e4bf51
 
 
 
 
ee0b265
 
068743a
ee0b265
07c92ca
 
 
 
 
 
 
 
 
 
 
 
 
 
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
---
language:
- en
license: cc-by-nc-4.0
datasets:
- c-s-ale/alpaca-gpt4-data
- Open-Orca/OpenOrca
- Intel/orca_dpo_pairs
- allenai/ultrafeedback_binarized_cleaned
base_model:
- upstage/SOLAR-10.7B-v1.0
model-index:
- name: SOLAR-10.7B-Instruct-v1.0
  results:
  - task:
      type: text-generation
      name: Text Generation
    dataset:
      name: AI2 Reasoning Challenge (25-Shot)
      type: ai2_arc
      config: ARC-Challenge
      split: test
      args:
        num_few_shot: 25
    metrics:
    - type: acc_norm
      value: 71.08
      name: normalized accuracy
    source:
      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=upstage/SOLAR-10.7B-Instruct-v1.0
      name: Open LLM Leaderboard
  - task:
      type: text-generation
      name: Text Generation
    dataset:
      name: HellaSwag (10-Shot)
      type: hellaswag
      split: validation
      args:
        num_few_shot: 10
    metrics:
    - type: acc_norm
      value: 88.16
      name: normalized accuracy
    source:
      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=upstage/SOLAR-10.7B-Instruct-v1.0
      name: Open LLM Leaderboard
  - task:
      type: text-generation
      name: Text Generation
    dataset:
      name: MMLU (5-Shot)
      type: cais/mmlu
      config: all
      split: test
      args:
        num_few_shot: 5
    metrics:
    - type: acc
      value: 66.21
      name: accuracy
    source:
      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=upstage/SOLAR-10.7B-Instruct-v1.0
      name: Open LLM Leaderboard
  - task:
      type: text-generation
      name: Text Generation
    dataset:
      name: TruthfulQA (0-shot)
      type: truthful_qa
      config: multiple_choice
      split: validation
      args:
        num_few_shot: 0
    metrics:
    - type: mc2
      value: 71.43
    source:
      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=upstage/SOLAR-10.7B-Instruct-v1.0
      name: Open LLM Leaderboard
  - task:
      type: text-generation
      name: Text Generation
    dataset:
      name: Winogrande (5-shot)
      type: winogrande
      config: winogrande_xl
      split: validation
      args:
        num_few_shot: 5
    metrics:
    - type: acc
      value: 83.58
      name: accuracy
    source:
      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=upstage/SOLAR-10.7B-Instruct-v1.0
      name: Open LLM Leaderboard
  - task:
      type: text-generation
      name: Text Generation
    dataset:
      name: GSM8k (5-shot)
      type: gsm8k
      config: main
      split: test
      args:
        num_few_shot: 5
    metrics:
    - type: acc
      value: 64.75
      name: accuracy
    source:
      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=upstage/SOLAR-10.7B-Instruct-v1.0
      name: Open LLM Leaderboard
---

<p align="left">
  <a href="https://go.upstage.ai/solar-obt-hf-modelcardv1-instruct">
    <img src="https://huggingface.co/upstage/SOLAR-10.7B-Instruct-v1.0/resolve/main/solar-api-banner.png" width="100%"/>
  </a>
<p>

# **Meet 10.7B Solar: Elevating Performance with Upstage Depth UP Scaling!**

**(This model is [upstage/SOLAR-10.7B-v1.0](https://huggingface.co/upstage/SOLAR-10.7B-v1.0) fine-tuned version for single-turn conversation.)**


# **Introduction**
We introduce SOLAR-10.7B, an advanced large language model (LLM) with 10.7 billion parameters, demonstrating superior performance in various natural language processing (NLP) tasks. It's compact, yet remarkably powerful, and demonstrates unparalleled state-of-the-art performance in models with parameters under 30B.

We present a methodology for scaling LLMs called depth up-scaling (DUS) , which encompasses architectural modifications and continued pretraining. In other words, we integrated Mistral 7B weights into the upscaled layers, and finally, continued pre-training for the entire model.


SOLAR-10.7B has remarkable performance. It outperforms models with up to 30B parameters, even surpassing the recent Mixtral 8X7B model. For detailed information, please refer to the experimental table.
Solar 10.7B is an ideal choice for fine-tuning. SOLAR-10.7B offers robustness and adaptability for your fine-tuning needs. Our simple instruction fine-tuning using the SOLAR-10.7B pre-trained model yields significant performance improvements.

For full details of this model please read our [paper](https://arxiv.org/abs/2312.15166).


# **Instruction Fine-Tuning Strategy**

We utilize state-of-the-art instruction fine-tuning methods including supervised fine-tuning (SFT) and direct preference optimization (DPO) [1].

We used a mixture of the following datasets
- c-s-ale/alpaca-gpt4-data (SFT)
- Open-Orca/OpenOrca (SFT)
- in-house generated data utilizing Metamath [2] (SFT, DPO)
- Intel/orca_dpo_pairs (DPO)
- allenai/ultrafeedback_binarized_cleaned (DPO)

where we were careful of data contamination by not using GSM8K samples when generating data and filtering tasks when applicable via the following list.
```python
filtering_task_list = [
    'task228_arc_answer_generation_easy',
    'ai2_arc/ARC-Challenge:1.0.0',
    'ai2_arc/ARC-Easy:1.0.0',
    'task229_arc_answer_generation_hard',
    'hellaswag:1.1.0', 
    'task1389_hellaswag_completion',
    'cot_gsm8k',
    'cot_gsm8k_ii',
    'drop:2.0.0',
    'winogrande:1.1.0'
]
```

Using the datasets mentioned above, we applied SFT and iterative DPO training, a proprietary alignment strategy, to maximize the performance of our resulting model.

[1] Rafailov, R., Sharma, A., Mitchell, E., Ermon, S., Manning, C.D. and Finn, C., 2023. Direct preference optimization: Your language model is secretly a reward model. NeurIPS.

[2] Yu, L., Jiang, W., Shi, H., Yu, J., Liu, Z., Zhang, Y., Kwok, J.T., Li, Z., Weller, A. and Liu, W., 2023. Metamath: Bootstrap your own mathematical questions for large language models. arXiv preprint arXiv:2309.12284.

# **Data Contamination Test Results**

Recently, there have been contamination issues in some models on the LLM leaderboard. 
We note that we made every effort to exclude any benchmark-related datasets from training.
We also ensured the integrity of our model by conducting a data contamination test [3] that is also used by the HuggingFace team [4, 5].

Our results, with `result < 0.1, %:` being well below 0.9, indicate that our model is free from contamination.

*The data contamination test results of HellaSwag and Winograde will be added once [3] supports them.*

| Model                        | ARC   | MMLU | TruthfulQA | GSM8K |
|------------------------------|-------|-------|-------|-------|
| **SOLAR-10.7B-Instruct-v1.0**| result < 0.1, %: 0.06 |result < 0.1, %: 0.15 | result < 0.1, %: 0.28 | result < 0.1, %: 0.70 |

[3] https://github.com/swj0419/detect-pretrain-code-contamination

[4] https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard/discussions/474#657f2245365456e362412a06

[5] https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard/discussions/265#657b6debf81f6b44b8966230

# **Evaluation Results**

| Model                                  | H6    | Model Size |
|----------------------------------------|-------|------------|
| **SOLAR-10.7B-Instruct-v1.0**              | **74.20** | **~ 11B**      |
| mistralai/Mixtral-8x7B-Instruct-v0.1   | 72.62 | ~ 46.7B    |
| 01-ai/Yi-34B-200K                      | 70.81 | ~ 34B      |
| 01-ai/Yi-34B                           | 69.42 | ~ 34B      |
| mistralai/Mixtral-8x7B-v0.1            | 68.42 | ~ 46.7B    |
| meta-llama/Llama-2-70b-hf              | 67.87 | ~ 70B      |
| tiiuae/falcon-180B                     | 67.85 | ~ 180B     |
| **SOLAR-10.7B-v1.0**                   | **66.04** | **~11B**   |
| mistralai/Mistral-7B-Instruct-v0.2     | 65.71 | ~ 7B       |
| Qwen/Qwen-14B                          | 65.86 | ~ 14B      |
| 01-ai/Yi-34B-Chat                      | 65.32 | ~34B       |
| meta-llama/Llama-2-70b-chat-hf         | 62.4  | ~ 70B      |
| mistralai/Mistral-7B-v0.1              | 60.97 | ~ 7B       |
| mistralai/Mistral-7B-Instruct-v0.1     | 54.96 | ~ 7B       |

# **Usage Instructions**

This model has been fine-tuned primarily for single-turn conversation, making it less suitable for multi-turn conversations such as chat.

### **Version**

Make sure you have the correct version of the transformers library installed:

```sh
pip install transformers==4.35.2
```

### **Loading the Model**

Use the following Python code to load the model:

```python
import torch
from transformers import AutoModelForCausalLM, AutoTokenizer

tokenizer = AutoTokenizer.from_pretrained("Upstage/SOLAR-10.7B-Instruct-v1.0")
model = AutoModelForCausalLM.from_pretrained(
    "Upstage/SOLAR-10.7B-Instruct-v1.0",
    device_map="auto",
    torch_dtype=torch.float16,
)
```

### **Conducting Single-Turn Conversation**

```python
conversation = [ {'role': 'user', 'content': 'Hello?'} ] 

prompt = tokenizer.apply_chat_template(conversation, tokenize=False, add_generation_prompt=True)

inputs = tokenizer(prompt, return_tensors="pt").to(model.device) 
outputs = model.generate(**inputs, use_cache=True, max_length=4096)
output_text = tokenizer.decode(outputs[0]) 
print(output_text)
```

Below is an example of the output.
```
<s> ### User:
Hello?

### Assistant:
Hello, how can I assist you today? Please feel free to ask any questions or request help with a specific task.</s>
```

### **License**
- [upstage/SOLAR-10.7B-v1.0](https://huggingface.co/upstage/SOLAR-10.7B-v1.0): apache-2.0
- [upstage/SOLAR-10.7B-Instruct-v1.0](https://huggingface.co/upstage/SOLAR-10.7B-Instruct-v1.0): cc-by-nc-4.0
  - Since some non-commercial datasets such as Alpaca are used for fine-tuning, we release this model as cc-by-nc-4.0.

### **How to Cite**

Please cite this model using this format.

```bibtex
@misc{kim2023solar,
      title={SOLAR 10.7B: Scaling Large Language Models with Simple yet Effective Depth Up-Scaling}, 
      author={Dahyun Kim and Chanjun Park and Sanghoon Kim and Wonsung Lee and Wonho Song and Yunsu Kim and Hyeonwoo Kim and Yungi Kim and Hyeonju Lee and Jihoo Kim and Changbae Ahn and Seonghoon Yang and Sukyung Lee and Hyunbyung Park and Gyoungjin Gim and Mikyoung Cha and Hwalsuk Lee and Sunghun Kim},
      year={2023},
      eprint={2312.15166},
      archivePrefix={arXiv},
      primaryClass={cs.CL}
}
```

### **The Upstage AI Team** ###
Upstage is creating the best LLM and DocAI. Please find more information at https://upstage.ai 

### **Contact Us** ###
Any questions and suggestions, please use the discussion tab. If you want to contact us directly, drop an email to [contact@upstage.ai](mailto:contact@upstage.ai)
# [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_upstage__SOLAR-10.7B-Instruct-v1.0)

|             Metric              |Value|
|---------------------------------|----:|
|Avg.                             |74.20|
|AI2 Reasoning Challenge (25-Shot)|71.08|
|HellaSwag (10-Shot)              |88.16|
|MMLU (5-Shot)                    |66.21|
|TruthfulQA (0-shot)              |71.43|
|Winogrande (5-shot)              |83.58|
|GSM8k (5-shot)                   |64.75|