Hugging Face's logo

nayohan
/
llama3-instrucTrans-enko-8b

Text Generation
Transformers
Safetensors
English
Korean
llama
translation
enko
conversational
text-generation-inference
Model card Files
xet
Community
5
nayohan commited on
Commit
101d48e
ยท
verified ยท
1 Parent(s): 4fd0b3e

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +191 -196
README.md CHANGED
@@ -1,199 +1,194 @@
1
  ---
 
 
 
 
2
  library_name: transformers
3
- tags: []
 
 
 
 
 
 
 
 
 
 
 
4
  ---
5
-
6
- # Model Card for Model ID
7
-
8
- <!-- Provide a quick summary of what the model is/does. -->
9
-
10
-
11
-
12
- ## Model Details
13
-
14
- ### Model Description
15
-
16
- <!-- Provide a longer summary of what this model is. -->
17
-
18
- This is the model card of a ๐Ÿค— transformers model that has been pushed on the Hub. This model card has been automatically generated.
19
-
20
- - **Developed by:** [More Information Needed]
21
- - **Funded by [optional]:** [More Information Needed]
22
- - **Shared by [optional]:** [More Information Needed]
23
- - **Model type:** [More Information Needed]
24
- - **Language(s) (NLP):** [More Information Needed]
25
- - **License:** [More Information Needed]
26
- - **Finetuned from model [optional]:** [More Information Needed]
27
-
28
- ### Model Sources [optional]
29
-
30
- <!-- Provide the basic links for the model. -->
31
-
32
- - **Repository:** [More Information Needed]
33
- - **Paper [optional]:** [More Information Needed]
34
- - **Demo [optional]:** [More Information Needed]
35
-
36
- ## Uses
37
-
38
- <!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
39
-
40
- ### Direct Use
41
-
42
- <!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
43
-
44
- [More Information Needed]
45
-
46
- ### Downstream Use [optional]
47
-
48
- <!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
49
-
50
- [More Information Needed]
51
-
52
- ### Out-of-Scope Use
53
-
54
- <!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
55
-
56
- [More Information Needed]
57
-
58
- ## Bias, Risks, and Limitations
59
-
60
- <!-- This section is meant to convey both technical and sociotechnical limitations. -->
61
-
62
- [More Information Needed]
63
-
64
- ### Recommendations
65
-
66
- <!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
67
-
68
- Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
69
-
70
- ## How to Get Started with the Model
71
-
72
- Use the code below to get started with the model.
73
-
74
- [More Information Needed]
75
-
76
- ## Training Details
77
-
78
- ### Training Data
79
-
80
- <!-- This should link to a Dataset 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. -->
81
-
82
- [More Information Needed]
83
-
84
- ### Training Procedure
85
-
86
- <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
87
-
88
- #### Preprocessing [optional]
89
-
90
- [More Information Needed]
91
-
92
-
93
- #### Training Hyperparameters
94
-
95
- - **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
96
-
97
- #### Speeds, Sizes, Times [optional]
98
-
99
- <!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
100
-
101
- [More Information Needed]
102
-
103
- ## Evaluation
104
-
105
- <!-- This section describes the evaluation protocols and provides the results. -->
106
-
107
- ### Testing Data, Factors & Metrics
108
-
109
- #### Testing Data
110
-
111
- <!-- This should link to a Dataset Card if possible. -->
112
-
113
- [More Information Needed]
114
-
115
- #### Factors
116
-
117
- <!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
118
-
119
- [More Information Needed]
120
-
121
- #### Metrics
122
-
123
- <!-- These are the evaluation metrics being used, ideally with a description of why. -->
124
-
125
- [More Information Needed]
126
-
127
- ### Results
128
-
129
- [More Information Needed]
130
-
131
- #### Summary
132
-
133
-
134
-
135
- ## Model Examination [optional]
136
-
137
- <!-- Relevant interpretability work for the model goes here -->
138
-
139
- [More Information Needed]
140
-
141
- ## Environmental Impact
142
-
143
- <!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
144
-
145
- Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
146
-
147
- - **Hardware Type:** [More Information Needed]
148
- - **Hours used:** [More Information Needed]
149
- - **Cloud Provider:** [More Information Needed]
150
- - **Compute Region:** [More Information Needed]
151
- - **Carbon Emitted:** [More Information Needed]
152
-
153
- ## Technical Specifications [optional]
154
-
155
- ### Model Architecture and Objective
156
-
157
- [More Information Needed]
158
-
159
- ### Compute Infrastructure
160
-
161
- [More Information Needed]
162
-
163
- #### Hardware
164
-
165
- [More Information Needed]
166
-
167
- #### Software
168
-
169
- [More Information Needed]
170
-
171
- ## Citation [optional]
172
-
173
- <!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
174
-
175
- **BibTeX:**
176
-
177
- [More Information Needed]
178
-
179
- **APA:**
180
-
181
- [More Information Needed]
182
-
183
- ## Glossary [optional]
184
-
185
- <!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
186
-
187
- [More Information Needed]
188
-
189
- ## More Information [optional]
190
-
191
- [More Information Needed]
192
-
193
- ## Model Card Authors [optional]
194
-
195
- [More Information Needed]
196
-
197
- ## Model Card Contact
198
-
199
- [More Information Needed]
 
1
  ---
2
+ language:
3
+ - en
4
+ - ko
5
+ license: llama3
6
  library_name: transformers
7
+ tags:
8
+ - translation
9
+ - enko
10
+ - ko
11
+ base_model:
12
+ - meta-llama/Meta-Llama-3-8B-Instruct
13
+ datasets:
14
+ - nayohan/aihub-en-ko-translation-1.2m
15
+ - nayohan/translate_corpus_313k
16
+ pipeline_tag: text-generation
17
+ metrics:
18
+ - sacrebleu
19
  ---
20
+ # **instructTrans**
21
+
22
+ ![image/png](https://cdn-uploads.huggingface.co/production/uploads/6152b4b9ecf3ca6ab820e325/oRlzxHQy3Qvqf4zfh5Wcj.png)
23
+ # **Introduction**
24
+
25
+ **llama3-8b-instructTrans-en-ko** model is trained on **translation datasets(english->korean)** based on Llama-3-8B-it.
26
+ - [nayohan/aihub-en-ko-translation-1.2m](https://huggingface.co/datasets/nayohan/aihub-en-ko-translation-1.2m)
27
+ - [nayohan/translate_corpus_313k](https://huggingface.co/datasets/nayohan/translate_corpus_313k)
28
+
29
+
30
+
31
+ ### **Loading the Model**
32
+ Use the following Python code to load the model:
33
+ ```python
34
+ import torch
35
+ from transformers import AutoModelForCausalLM, AutoTokenizer
36
+
37
+ model_name = "nayohan/llama3-instrucTrans-enko-8b"
38
+ tokenizer = AutoTokenizer.from_pretrained(model_name)
39
+ model = AutoModelForCausalLM.from_pretrained(
40
+ model_name,
41
+ device_map="auto",
42
+ torch_dtype=torch.bfloat16
43
+ )
44
+ ```
45
+
46
+ ### **Generating Text**
47
+ This model supports translation from english to korean. To translate text, use the following Python code:
48
+ ```python
49
+ system_prompt="๋‹น์‹ ์€ ๋ฒˆ์—ญ๊ธฐ ์ž…๋‹ˆ๋‹ค. ์˜์–ด๋ฅผ ํ•œ๊ตญ์–ด๋กœ ๋ฒˆ์—ญํ•˜์„ธ์š”."
50
+ sentence = "The aerospace industry is a flower in the field of technology and science."
51
+ conversation = [{'role': 'system', 'content': system_prompt},
52
+ {'role': 'user', 'content': sentence}]
53
+
54
+ inputs = tokenizer.apply_chat_template(
55
+ conversation,
56
+ tokenize=True,
57
+ add_generation_prompt=True,
58
+ return_tensors='pt'
59
+ ).to("cuda")
60
+
61
+ outputs = model.generate(inputs, max_new_tokens=4096) # Finetuned with length 4096
62
+ print(tokenizer.decode(outputs[0][len(inputs[0]):]))
63
+ ```
64
+ ```
65
+ # Result
66
+ INPUT: <|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\n๋‹น์‹ ์€ ๋ฒˆ์—ญ๊ธฐ ์ž…๋‹ˆ๋‹ค. ์˜์–ด๋ฅผ ํ•œ๊ตญ์–ด๋กœ ๋ฒˆ์—ญํ•˜์„ธ์š”.<|eot_id|><|start_header_id|>user<|end_header_id|>\n\nThe aerospace industry is a flower in the field of technology and science.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n
67
+ OUTPUT: ํ•ญ๊ณต์šฐ์ฃผ ์‚ฐ์—…์€ ๊ธฐ์ˆ ๊ณผ ๊ณผํ•™ ๋ถ„์•ผ์˜ ๊ฝƒ์ž…๋‹ˆ๋‹ค.<|eot_id|>
68
+
69
+ INPUT: <|begin_of_text|><|start_header_id|>system<|end_header_id|>\n\n๋‹น์‹ ์€ ๋ฒˆ์—ญ๊ธฐ ์ž…๋‹ˆ๋‹ค. ์˜์–ด๋ฅผ ํ•œ๊ตญ์–ด๋กœ ๋ฒˆ์—ญํ•˜์„ธ์š”.<|eot_id|><|start_header_id|>user<|end_header_id|>\n\n
70
+ Technical and basic sciences are very important in terms of research. It has a significant impact on the industrial development of a country. Government policies control the research budget.<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n
71
+ OUTPUT: ๊ธฐ์ˆ  ๋ฐ ๊ธฐ์ดˆ ๊ณผํ•™์€ ์—ฐ๊ตฌ ์ธก๋ฉด์—์„œ ๋งค์šฐ ์ค‘์š”ํ•ฉ๋‹ˆ๋‹ค. ์ด๋Š” ํ•œ ๊ตญ๊ฐ€์˜ ์‚ฐ์—… ๋ฐœ์ „์— ํฐ ์˜ํ–ฅ์„ ๋ฏธ์นฉ๋‹ˆ๋‹ค. ์ •๋ถ€ ์ •์ฑ…์€ ์—ฐ๊ตฌ ์˜ˆ์‚ฐ์„ ํ†ต์ œํ•ฉ๋‹ˆ๋‹ค.<|eot_id|>
72
+ ```
73
+ ```
74
+ # EVAL_RESULT (2405_KO_NEWS) (max_new_tokens=512)
75
+ "en_ref":"This controversy arose around a new advertisement for the latest iPad Pro that Apple released on YouTube on the 7th. The ad shows musical instruments, statues, cameras, and paints being crushed in a press, followed by the appearance of the iPad Pro in their place. It appears to emphasize the new iPad Pro's artificial intelligence features, advanced display, performance, and thickness. Apple mentioned that the newly unveiled iPad Pro is equipped with the latest 'M4' chip and is the thinnest device in Apple's history. The ad faced immediate backlash upon release, as it graphically depicts objects symbolizing creators being crushed. Critics argue that the imagery could be interpreted as technology trampling on human creators. Some have also voiced concerns that it evokes a situation where creators are losing ground due to AI."
76
+ "ko_ref":"์ด๋ฒˆ ๋…ผ๋ž€์€ ์• ํ”Œ์ด ์ง€๋‚œ 7์ผ ์œ ํŠœ๋ธŒ์— ๊ณต๊ฐœํ•œ ์‹ ํ˜• ์•„์ดํŒจ๋“œ ํ”„๋กœ ๊ด‘๊ณ ๋ฅผ ๋‘˜๋Ÿฌ์‹ธ๊ณ  ๋ถˆ๊ฑฐ์กŒ๋‹ค. ํ•ด๋‹น ๊ด‘๊ณ  ์˜์ƒ์€ ์•…๊ธฐ์™€ ์กฐ๊ฐ์ƒ, ์นด๋ฉ”๋ผ, ๋ฌผ๊ฐ ๋“ฑ์„ ์••์ฐฉ๊ธฐ๋กœ ์ง“๋ˆ„๋ฅธ ๋’ค ๊ทธ ์ž๋ฆฌ์— ์•„์ดํŒจ๋“œ ํ”„๋กœ๋ฅผ ๋“ฑ์žฅ์‹œํ‚ค๋Š” ๋‚ด์šฉ์ด์—ˆ๋‹ค. ์‹ ํ˜• ์•„์ดํŒจ๋“œ ํ”„๋กœ์˜ ์ธ๊ณต์ง€๋Šฅ ๊ธฐ๋Šฅ๋“ค๊ณผ ์ง„ํ™”๋œ ๋””์Šคํ”Œ๋ ˆ์ด์™€ ์„ฑ๋Šฅ, ๋‘๊ป˜ ๋“ฑ์„ ๊ฐ•์กฐํ•˜๊ธฐ ์œ„ํ•œ ์ทจ์ง€๋กœ ํ’€์ด๋œ๋‹ค. ์• ํ”Œ์€ ์ด๋ฒˆ์— ๊ณต๊ฐœํ•œ ์•„์ดํŒจ๋“œ ํ”„๋กœ์— ์‹ ํ˜• โ€˜M4โ€™ ์นฉ์ด ํƒ‘์žฌ๋˜๋ฉฐ ๋‘๊ป˜๋Š” ์• ํ”Œ์˜ ์—ญ๋Œ€ ์ œํ’ˆ ์ค‘ ๊ฐ€์žฅ ์–‡๋‹ค๋Š” ์„ค๋ช…๋„ ๋ง๋ถ™์˜€๋‹ค. ๊ด‘๊ณ ๋Š” ๊ณต๊ฐœ ์งํ›„ ๊ฑฐ์„ผ ๋น„ํŒ์— ์ง๋ฉดํ–ˆ๋‹ค. ์ฐฝ์ž‘์ž๋ฅผ ์ƒ์ง•ํ•˜๋Š” ๋ฌผ๊ฑด์ด ์ง“๋ˆŒ๋ ค์ง€๋Š” ๊ณผ์ •์„ ์ง€๋‚˜์น˜๊ฒŒ ์ ๋‚˜๋ผํ•˜๊ฒŒ ๋ฌ˜์‚ฌํ•œ ์ ์ด ๋ฌธ์ œ๊ฐ€ ๋๋‹ค. ๊ธฐ์ˆ ์ด ์ธ๊ฐ„ ์ฐฝ์ž‘์ž๋ฅผ ์ง“๋ฐŸ๋Š” ๋ชจ์Šต์„ ๋ฌ˜์‚ฌํ•œ ๊ฒƒ์œผ๋กœ ํ•ด์„๋  ์—ฌ์ง€๊ฐ€ ์žˆ๋‹ค๋Š” ๋ฌธ์ œ์˜์‹์ด๋‹ค. ์ธ๊ณต์ง€๋Šฅ(AI)์œผ๋กœ ์ธํ•ด ์ฐฝ์ž‘์ž๊ฐ€ ์„ค ์ž๋ฆฌ๊ฐ€ ์ค„์–ด๋“œ๋Š” ์ƒํ™ฉ์„ ์—ฐ์ƒ์‹œํ‚จ๋‹ค๋Š” ๋ชฉ์†Œ๋ฆฌ๋„ ๋‚˜์™”๋‹ค."
77
+
78
+ "InstrucTrans":"์ด๋ฒˆ ๋…ผ๋ž€์€ ์• ํ”Œ์ด ์ง€๋‚œ 7์ผ ์œ ํŠœ๋ธŒ์— ๊ณต๊ฐœํ•œ ์ตœ์‹  ์•„์ดํŒจ๋“œ ํ”„๋กœ ๊ด‘๊ณ ๋ฅผ ์ค‘์‹ฌ์œผ๋กœ ๋ถˆ๊ฑฐ์กŒ๋‹ค. ์ด ๊ด‘๊ณ ๋Š” ์•…๊ธฐ, ์กฐ๊ฐ์ƒ, ์นด๋ฉ”๋ผ, ๋ฌผ๊ฐ ๋“ฑ์„ ๋ˆ„๋ฅด๊ธฐ ์‹œ์ž‘ํ•˜๋Š” ์žฅ๋ฉด๊ณผ ํ•จ๊ป˜ ๊ทธ ์ž๋ฆฌ์— ์•„์ดํŒจ๋“œ ํ”„๋กœ๊ฐ€ ๋“ฑ์žฅํ•˜๋Š” ์žฅ๋ฉด์„ ๋ณด์—ฌ์ค€๋‹ค. ์ด๋Š” ์ƒˆ๋กœ์šด ์•„์ดํŒจ๋“œ ํ”„๋กœ์˜ ์ธ๊ณต์ง€๋Šฅ ๊ธฐ๋Šฅ, ๊ณ ๊ธ‰ ๋””์Šคํ”Œ๋ ˆ์ด, ์„ฑ๋Šฅ, ๋‘๊ป˜๋ฅผ ๊ฐ•์กฐํ•˜๋Š” ๊ฒƒ์œผ๋กœ ๋ณด์ธ๋‹ค. ์• ํ”Œ์€ ์ด๋ฒˆ์— ๊ณต๊ฐœํ•œ ์•„์ดํŒจ๋“œ ํ”„๋กœ์— ์ตœ์‹  'M4' ์นฉ์ด ํƒ‘์žฌ๋์œผ๋ฉฐ, ์• ํ”Œ ์—ญ์‚ฌ์ƒ ๊ฐ€์žฅ ์–‡์€ ๊ธฐ๊ธฐ๋ผ๊ณ  ์–ธ๊ธ‰ํ–ˆ๋‹ค. ์ด ๊ด‘๊ณ ๋Š” ์ถœ์‹œํ•˜์ž๋งˆ์ž ํฌ๋ฆฌ์—์ดํ„ฐ๋ฅผ ์ƒ์ง•ํ•˜๋Š” ๋ฌผ๊ฑด์ด ํŒŒ์‡„๋˜๋Š” ์žฅ๋ฉด์ด ๊ทธ๋Œ€๋กœ ๊ทธ๋ ค์ ธ ๋…ผ๋ž€์ด ๋˜๊ณ  ์žˆ๋‹ค. ๋น„ํ‰๊ฐ€๋“ค์€ ์ด ์ด๋ฏธ์ง€๊ฐ€ ๊ธฐ์ˆ ์ด ์ธ๊ฐ„ ํฌ๋ฆฌ์—์ดํ„ฐ๋ฅผ ์ง“๋ฐŸ๋Š”๋‹ค๋Š” ์˜๋ฏธ๋กœ ํ•ด์„๋  ์ˆ˜ ์žˆ๋‹ค๊ณ  ์ฃผ์žฅํ•œ๋‹ค. ๋˜ํ•œ AI๋กœ ์ธํ•ด ํฌ๋ฆฌ์—์ดํ„ฐ๋“ค์ด ๋ฐ€๋ฆฌ๊ณ  ์žˆ๋‹ค๋Š” ์ƒํ™ฉ์„ ์—ฐ์ƒ์‹œํ‚จ๋‹ค๋Š” ์šฐ๋ ค์˜ ๋ชฉ์†Œ๋ฆฌ๋„ ๋‚˜์˜จ๋‹ค."
79
+
80
+ "KULLM3":"์ด ๋…ผ๋ž€์€ ์• ํ”Œ์ด ์œ ํŠœ๋ธŒ์— 7์ผ์— ๋ฐœํ‘œํ•œ ์ตœ์‹  iPad Pro ๊ด‘๊ณ  ์ฃผ์œ„์—์„œ ๋ฐœ์ƒํ–ˆ์Šต๋‹ˆ๋‹ค. ๊ด‘๊ณ ์—์„œ๋Š” ์•…๊ธฐ, ์กฐ๊ฐ์ƒ, ์นด๋ฉ”๋ผ, ๊ทธ๋ฆฌ๊ณ  ๋ฌผ๊ฐ์ด ์••์ถ•๊ธฐ์—์„œ ํŒŒ๊ดด๋˜๋Š” ๋ชจ์Šต์ด ๋ณด์—ฌ์ง€๊ณ , ๊ทธ ์ž๋ฆฌ์— iPad Pro๊ฐ€ ๋‚˜ํƒ€๋‚ฉ๋‹ˆ๋‹ค. ์ด๋Š” ์ƒˆ๋กœ์šด iPad Pro์˜ ์ธ๊ณต์ง€๋Šฅ ๊ธฐ๋Šฅ, ๊ณ ๊ธ‰ ๋””์Šคํ”Œ๋ ˆ์ด, ์„ฑ๋Šฅ, ๊ทธ๋ฆฌ๊ณ  ์–‡์€ ๋””์ž์ธ์„ ๊ฐ•์กฐํ•˜๋Š” ๊ฒƒ์œผ๋กœ ๋ณด์ž…๋‹ˆ๋‹ค. ์• ํ”Œ์€ ์ตœ์‹  'M4' ์นฉ์„ ํƒ‘์žฌํ•œ ์ƒˆ๋กœ ๊ณต๊ฐœ๋œ iPad Pro๊ฐ€ ์ž์‚ฌ ์—ญ์‚ฌ์ƒ ๊ฐ€์žฅ ์–‡์€ ๊ธฐ๊ธฐ๋ผ๊ณ  ์–ธ๊ธ‰ํ–ˆ์Šต๋‹ˆ๋‹ค. ํ•˜์ง€๋งŒ ๊ด‘๊ณ ๋Š” ์ถœ์‹œ ์งํ›„ ์ฆ‰๊ฐ์ ์ธ ๋ฐ˜๋ฐœ์„ ๋ฐ›์•˜์Šต๋‹ˆ๋‹ค. ๊ด‘๊ณ ์—์„œ๋Š” ์ฐฝ์ž‘์ž๋ฅผ ์ƒ์ง•ํ•˜๋Š” ๋ฌผ๊ฑด๋“ค์ด ํŒŒ๊ดด๋˜๋Š” ๋ชจ์Šต์ด ๊ทธ๋ž˜ํ”ฝํ•˜๊ฒŒ ๋ณด์—ฌ์ง€๊ธฐ ๋•Œ๋ฌธ์ž…๋‹ˆ๋‹ค. ๋น„ํŒ์ž๋“ค์€ ์ด ์ด๋ฏธ์ง€๊ฐ€ ๊ธฐ์ˆ ์ด ์ธ๊ฐ„ ์ฐฝ์ž‘์ž๋ฅผ ์••๋„ํ•˜๋Š” ๊ฒƒ์œผ๋กœ ํ•ด์„๋  ์ˆ˜ ์žˆ๋‹ค๊ณ  ์ฃผ์žฅํ•˜๋ฉฐ, ์ผ๋ถ€๋Š” ์ด๊ฐ€ ์ฐฝ์ž‘์ž๋“ค์ด AI ๋•Œ๋ฌธ์— ์ง€์œ„๋ฅผ ์žƒ๋Š” ์ƒํ™ฉ์„ ์—ฐ์ƒ์‹œํ‚จ๋‹ค๊ณ  ์šฐ๋ คํ•˜๊ณ  ์žˆ์Šต๋‹ˆ๋‹ค."
81
+ "EEVE-10.8b-it":ํ•ด๋‹น ๋…ผ๋ž€์€ ์• ํ”Œ์ด 7์ผ์— ์œ ํŠœ๋ธŒ์— ๊ณต๊ฐœํ•œ ์ตœ์‹  ์•„์ดํŒจ๋“œ ํ”„๋กœ ๊ด‘๊ณ ์™€ ๊ด€๋ จํ•˜์—ฌ ๋ฐœ์ƒํ–ˆ์Šต๋‹ˆ๋‹ค. ํ•ด๋‹น ๊ด‘๊ณ ์—์„œ๋Š” ์•…๊ธฐ, ์กฐ๊ฐ์ƒ, ์นด๋ฉ”๋ผ, ๊ทธ๋ฆฌ๊ณ  ๋ถ“์ด ๋ˆŒ๋Ÿฌ์ ธ ๋ถ€์„œ์ง€๋Š” ๋ชจ์Šต๊ณผ ํ•จ๊ป˜ ๊ทธ ์ž๋ฆฌ์— ์ƒˆ๋กœ์šด ์•„์ดํŒจ๋“œ ํ”„๋กœ๊ฐ€ ๋“ฑ์žฅํ•˜๋Š” ์žฅ๋ฉด์„ ์ƒ์ƒํ•˜๊ฒŒ ๋ณด์—ฌ์ฃผ๊ณ  ์žˆ์Šต๋‹ˆ๋‹ค. ์ด๋Š” ์ƒˆ๋กœ์šด ์•„์ดํŒจ๋“œ ํ”„๋กœ์˜ ์ธ๊ณต์ง€๋Šฅ ๊ธฐ๋Šฅ, ์ง„๋ณด๋œ ๋””์Šคํ”Œ๋ ˆ์ด, ์„ฑ๋Šฅ, ๊ทธ๋ฆฌ๊ณ  ๋‘๊ป˜๋ฅผ ๋ถ€๊ฐ์‹œํ‚ค๊ณ ์ž ํ•˜๋Š” ๊ฒƒ์œผ๋กœ ๋ณด์ž…๋‹ˆ๋‹ค. ์• ํ”Œ์€ ๊ฐ“ ๋ฐœํ‘œ๋œ ์•„์ดํŒจ๋“œ ํ”„๋กœ๊ฐ€ ์ตœ์‹  'M4' ์นฉ์„ ํƒ‘์žฌํ•˜๊ณ  ์žˆ์œผ๋ฉฐ, ์• ํ”Œ ์ œํ’ˆ ์ค‘ ๊ฐ€์žฅ ์–‡์€ ์žฅ์น˜๋ผ๊ณ  ์–ธ๊ธ‰ํ–ˆ์Šต๋‹ˆ๋‹ค. ๊ด‘๊ณ ๊ฐ€ ์ถœ์‹œ๋˜์ž๋งˆ์ž ๋ฐ”๋กœ ๋ญ‡๋งค๋ฅผ ๋งž์•˜๋Š”๋ฐ, ์ด๋Š” ์ฐฝ์ž‘์ž๋“ค์„ ์ƒ์ง•ํ•˜๋Š” ๋ฌผ๊ฑด๋“ค์ด ๋ˆŒ๋ ค ๋ถ€์„œ์ง€๋Š” ์ž”์ธํ•œ ์žฅ๋ฉด์„ ๋‹ด๊ณ  ์žˆ๊ธฐ ๋•Œ๋ฌธ์ž…๋‹ˆ๋‹ค. ๋น„ํŒ์ž๋“ค์€ ์ด๋Ÿฌํ•œ ์ด๋ฏธ์ง€๊ฐ€ ๊ธฐ์ˆ ์ด ์ธ๊ฐ„ ์ฐฝ์ž‘์ž๋“ค์„ ์ง“๋ฐŸ์€ ๊ฒƒ์œผ๋กœ ํ•ด์„๋  ์ˆ˜ ์žˆ๋‹ค๊ณ  ์ฃผ์žฅํ•˜๊ณ  ์žˆ์Šต๋‹ˆ๋‹ค. ์ผ๋ถ€ ์‚ฌ๋žŒ๋“ค์€ ๋˜ํ•œ AI๋กœ ์ธํ•ด ์ฐฝ์ž‘์ž๋“ค์ด ๋„ํƒœ๋˜๊ณ  ์žˆ๋Š” ์ƒํ™ฉ์„ ์•”์‹œํ•˜๋Š” ๊ฒƒ ๊ฐ™์•„ ์šฐ๋ ค๋ฅผ ํ‘œํ•˜๊ณ  ์žˆ์Šต๋‹ˆ๋‹ค."
82
+ "Seagull-13B":"์ด ๋…ผ๋ž€์ด ๋ฐœ์ƒํ•œ ๊ฒƒ์€ 7์ผ์— Apple์ด YouTube์— ๊ณต๊ฐœํ•œ ์ตœ์‹  iPad Pro์˜ ์ƒˆ๋กœ์šด ๊ด‘๊ณ ์™€ ๊ด€๋ จ์ด ์žˆ์Šต๋‹ˆ๋‹ค. ์ด ๊ด‘๊ณ ๋Š” ์•…๊ธฐ, ๋™์ƒ, ์นด๋ฉ”๋ผ, ๋ฌผ๊ฐ์„ ํ”„๋ ˆ์Šค์— ๋ˆ„๋ฅด๋Š” ์žฅ๋ฉด์„ ๋ณด์—ฌ์ค€ ๋‹ค์Œ ๊ทธ ์ž๋ฆฌ์— iPad Pro๊ฐ€ ๋‚˜ํƒ€๋‚ฉ๋‹ˆ๋‹ค. ์ƒˆ๋กœ์šด iPad Pro์˜ ์ธ๊ณต์ง€๋Šฅ ๊ธฐ๋Šฅ, ๊ณ ๊ธ‰ ๋””์Šคํ”Œ๋ ˆ์ด, ์„ฑ๋Šฅ ๋ฐ ๋‘๊ป˜๋ฅผ ๊ฐ•์กฐํ•˜๋Š” ๊ฒƒ์œผ๋กœ ๋ณด์ž…๋‹ˆ๋‹ค. Apple์€ ์ƒˆ๋กœ ๊ณต๊ฐœ ๋œ iPad Pro์— ์ตœ์‹  'M4' ์นฉ์ด ์žฅ์ฐฉ๋˜์–ด ์žˆ์œผ๋ฉฐ Apple ์—ญ์‚ฌ์ƒ ๊ฐ€์žฅ ์–‡์€ ์žฅ์น˜๋ผ๊ณ  ์–ธ๊ธ‰ํ–ˆ์Šต๋‹ˆ๋‹ค. ์ด ๊ด‘๊ณ ๋Š” ํฌ๋ฆฌ์—์ดํ„ฐ๋ฅผ ์ƒ์ง•ํ•˜๋Š” ๋ฌผ์ฒด๋ฅผ ๊ทธ๋ž˜ํ”ฝ์œผ๋กœ ํ‘œํ˜„ํ•˜์—ฌ ๊ณต๊ฐœ ์งํ›„ ๋ฐ˜๋ฐœ์„ ๋ถˆ๋Ÿฌ์ผ์œผ์ผฐ์Šต๋‹ˆ๋‹ค. ๋น„ํ‰๊ฐ€๋“ค์€ ์ด ์ด๋ฏธ์ง€๊ฐ€ ๋”ฅ๋Ÿฌ๋‹์ด ์ธ๊ฐ„ ์ฐฝ์ž‘์ž๋ฅผ ์ง“๋ฐŸ๋Š” ๊ธฐ์ˆ ๋กœ ํ•ด์„๋  ์ˆ˜ ์žˆ๋‹ค๊ณ  ์ง€์ ํ•ฉ๋‹ˆ๋‹ค. ์ผ๋ถ€๋Š” AI๋กœ ์ธํ•ด ํฌ๋ฆฌ์—์ดํ„ฐ๊ฐ€ ํ† ๋Œ€๋ฅผ ์žƒ๋Š” ์ƒํ™ฉ์„ ์—ฐ์ƒ์‹œํ‚จ๋‹ค๋Š” ์šฐ๋ ค์˜ ๋ชฉ์†Œ๋ฆฌ๋„ ์ œ๊ธฐํ–ˆ์Šต๋‹ˆ๋‹ค.\n"
83
+ "Synatra-7B":"์ด ๋…ผ๋ž€์€ 7์ผ Apple์ด YouTube์— ์˜ฌ๋ฆฐ ์ตœ์‹  iPad Pro์˜ ์‹ ๊ทœ ๊ด‘๊ณ ๋ฅผ ์ค‘์‹ฌ์œผ๋กœ ๋ฒŒ์–ด์กŒ์Šต๋‹ˆ๋‹ค. ์ด ๊ด‘๊ณ ์—๋Š” ๋ชฉ๊ด€์•…๊ธฐ, ์กฐ๊ฐ์ƒ, ์นด๋ฉ”๋ผ, ๋ฌผ๊ฐ์ด ์••๋ ฅ์„ ๊ฐ€ํ•œ ํ›„ ๊ทธ ์ž๋ฆฌ์— iPad Pro๊ฐ€ ๋‚˜ํƒ€๋‚˜๋Š” ์žฅ๋ฉด์ด ํ‘œ์‹œ๋˜๋Š” ๋ฌด๋‹จ์žฅ์‹ ๊ด‘๊ณ ์ž…๋‹ˆ๋‹ค. ๊ทธ ๊ด‘๊ณ ๋Š” ์ƒˆ iPad Pro์˜ ์ธ๊ณต์ง€๋Šฅ ๊ธฐ๋Šฅ, ๊ณ ๊ธ‰ ๋””์Šคํ”Œ๋ ˆ์ด, ์„ฑ๋Šฅ ๋ฐ ๋‘๊ป˜๋ฅผ ๊ฐ•์กฐํ•˜๋Š” ๊ฒƒ์œผ๋กœ ๋ณด์ž…๋‹ˆ๋‹ค. Apple์€ ์ƒˆ๋กญ๊ฒŒ ๋ฐœํ‘œ๋œ iPad Pro์—๋Š” ์ตœ์‹  'M4' ์นฉ์ด ํƒ‘์žฌ๋˜์–ด ์žˆ์œผ๋ฉฐ ์—ญ์‚ฌ์ƒ ๊ฐ€์žฅ ์–‡์€ ๊ธฐ๊ธฐ๋ผ๊ณ  ์–ธ๊ธ‰ํ–ˆ์Šต๋‹ˆ๋‹ค. ์ด ๊ด‘๊ณ ๋Š” ํฌ๋ฆฌ์—์ดํ„ฐ๋ฅผ ์ƒ์ง•ํ•˜๋Š” ๋ฌผ์ฒด๊ฐ€ ์ง“๊ธฐ์— ๋งž์„œ ์žˆ๋‹ค๋Š” ๋ชจ์Šต์„ ๊ทธ๋ž˜ํ”ฝ์œผ๋กœ ํ‘œํ˜„ํ•œ ํ›„ ์ฆ‰์‹œ ๋ฐ˜๋ฐœ์„ ๋ถˆ๋Ÿฌ ์ผ์œผ์ผฐ์Šต๋‹ˆ๋‹ค. ๋น„ํ‰๊ฐ€๋“ค์€ ์ด ์ด๋ฏธ์ง€๊ฐ€ ๊ธฐ์ˆ ์ด ์ธ๊ฐ„ ํฌ๋ฆฌ์—์ดํ„ฐ๋ฅผ ํ˜‘๋ฐ•ํ•˜๋Š” ๊ฒƒ์œผ๋กœ ํ•ด์„๋  ์ˆ˜ ์žˆ๋‹ค๊ณ  ์ฃผ์žฅํ•ฉ๋‹ˆ๏ฟฝ๏ฟฝ. ์ผ๋ถ€๋Š” ๋˜ํ•œ AI๋กœ ์ธํ•ด ํฌ๋ฆฌ์—์ดํ„ฐ๊ฐ€ ์ง€์œ„๋ฅผ ์žƒ๋Š” ์ƒํ™ฉ์„ ๋ถˆ๋Ÿฌ์ผ์œผํ‚ฌ ์ˆ˜ ์žˆ๋‹ค๊ณ  ์šฐ๋ คํ•˜๋Š” ๋ชฉ์†Œ๋ฆฌ๋„ ์žˆ์Šต๋‹ˆ๋‹ค."
84
+ "nhndq-nllb":"์ด ๋…ผ๋ž€์€ ์• ํ”Œ์ด 7์ผ ์œ ํŠœ๋ธŒ์— ๊ณต๊ฐœํ•œ ์ตœ์‹  ์•„์ดํŒจ๋“œ ํ”„๋กœ์˜ ์ƒˆ ๊ด‘๊ณ ๋ฅผ ๋‘˜๋Ÿฌ์‹ธ๊ณ  ๋ถˆ๊ฑฐ์กŒ๋‹ค. ๊ด‘๊ณ ๋Š” ์•…๊ธฐ, ์กฐ๊ฐ์ƒ, ์นด๋ฉ”๋ผ, ํŽ˜์ธํŠธ ๋“ฑ์ด ํ”„๋ ˆ์Šค์—์„œ ์œผ๊นจ์ง€๊ณ  ๊ทธ ์ž๋ฆฌ์— ์•„์ดํŒจ๋“œ ํ”„๋กœ๊ฐ€ ๋“ฑ์žฅํ•˜๋Š” ๋ชจ์Šต์„ ๋ณด์—ฌ์ค€๋‹ค. ์ด๋Š” ์ƒˆ๋กœ์šด ์•„์ดํŒจ๋“œ ํ”„๋กœ์˜ ์ธ๊ณต์ง€๋Šฅ ๊ธฐ๋Šฅ๊ณผ ๊ณ ๊ธ‰ ๋””์Šคํ”Œ๋ ˆ์ด, ์„ฑ๋Šฅ, ๋‘๊ป˜ ๋“ฑ์„ ๊ฐ•์กฐํ•˜๋Š” ๊ฒƒ์œผ๋กœ ๋ณด์ธ๋‹ค. ์• ํ”Œ์€ ์ƒˆ๋กœ ๊ณต๊ฐœ๋œ ์•„์ดํŒจ๋“œ ํ”„๋กœ๊ฐ€ ์ตœ์‹  'M4' ์นฉ์„ ์žฅ์ฐฉํ•˜๊ณ  ์žˆ์œผ๋ฉฐ ์• ํ”Œ ์—ญ์‚ฌ์ƒ ๊ฐ€์žฅ ์–‡์€ ์žฅ์น˜๋ผ๊ณ  ์–ธ๊ธ‰ํ–ˆ๋‹ค. AI๋กœ ์ธํ•ด ์ฆ‰๊ฐ"
85
+
86
+ "our-tech":"์ด๋ฒˆ ๋…ผ๋ž€์€ ์• ํ”Œ์ด ์ง€๋‚œ 7์ผ ์œ ํŠœ๋ธŒ์— ๊ณต๊ฐœํ•œ ์ตœ์‹  ์•„์ดํŒจ๋“œ ํ”„๋กœ ๊ด‘๊ณ ๋ฅผ ๋‘˜๋Ÿฌ์‹ธ๊ณ  ๋ถˆ๊ฑฐ์กŒ๋‹ค. ๊ด‘๊ณ ๋Š” ์•…๊ธฐ, ์กฐ๊ฐ์ƒ, ์นด๋ฉ”๋ผ, ๋ฌผ๊ฐ ๋“ฑ์„ ์••์ฐฉ๊ธฐ์— ๋„ฃ์–ด ๋ถ€์ˆด๋ฒ„๋ฆฌ๋‹ค๊ฐ€ ๊ทธ ์ž๋ฆฌ์— ์•„์ดํŒจ๋“œ ํ”„๋กœ๊ฐ€ ๋“ฑ์žฅํ•˜๋Š” ๊ฒƒ์œผ๋กœ, ์ƒˆ๋กœ ๊ณต๊ฐœ๋œ ์•„์ดํŒจ๋“œ ํ”„๋กœ๊ฐ€ ์ตœ์‹  'M4'์นฉ์„ ํƒ‘์žฌํ•˜๊ณ  ์• ํ”Œ ์‚ฌ์ƒ ๊ฐ€์žฅ ์–‡์€ ๊ธฐ๊ธฐ๋ผ๋Š” ์ ์„ ๊ฐ•์กฐํ•œ ๊ฒƒ์œผ๋กœ ๋ณด์ธ๋‹ค. ๊ด‘๊ณ ๋Š” ์ถœ์‹œ ์ฆ‰์‹œ ์ฐฝ์ž‘์ž๋ฅผ ์ƒ์ง•ํ•˜๋Š” ๋ฌผ๊ฑด๋“ค์ด ์••์ฐฉ๊ธฐ์— ๊ฐˆ๊ฒจ๋ฒ„๋ฆฌ๋Š” ์žฅ๋ฉด์„ ๊ทธ๋ž˜ํ”ฝ์œผ๋กœ ๋ณด์—ฌ์ค˜, ๊ธฐ์ˆ ์ด ์ธ๊ฐ„ ์ฐฝ์ž‘์ž๋ฅผ ์ง“๋ฐŸ๋Š” ๊ฒƒ์œผ๋กœ ํ•ด์„๋  ์ˆ˜ ์žˆ๋‹ค๋Š” ์ง€์ ๊ณผ ํ•จ๊ป˜, AI๋กœ ์ธํ•ด ์ฐฝ์ž‘์ž๋“ค์ด ์ง€์œ„๋ฅผ ์žƒ์–ด๊ฐ€๋Š” ์ƒํ™ฉ์„ ์—ฐ์ƒ์‹œํ‚จ๋‹ค๋Š” ๋น„ํŒ์ด ์ œ๊ธฐ๋๋‹ค."
87
+ "our-general":์ด๋ฒˆ ๋…ผ๋ž€์€ ์• ํ”Œ์ด ์ง€๋‚œ 7์ผ ์œ ํŠœ๋ธŒ์— ๊ณต๊ฐœํ•œ ์ตœ์‹  ์•„์ดํŒจ๋“œ ํ”„๋กœ ๊ด‘๊ณ ๋ฅผ ์ค‘์‹ฌ์œผ๋กœ ๋ถˆ๊ฑฐ์กŒ๋‹ค. ๊ด‘๊ณ ๋Š” ์•…๊ธฐ, ์กฐ๊ฐ์ƒ, ์นด๋ฉ”๋ผ, ๋ฌผ๊ฐ ๋“ฑ์„ ๋ˆ„๋ฅด๊ธฐ์— ์ถฉ๋ถ„ํ•œ ํž˜์„ ๊ฐ€์ง„ ํ”„๋ ˆ์Šค์— ์ง‘์–ด๋„ฃ๊ณ  ์œผ๊นจ๋Š” ๋ชจ์Šต์„ ๋ณด์—ฌ์ค€๋‹ค. ์ด์–ด ๊ทธ ์ž๋ฆฌ์— ์•„์ดํŒจ๋“œ ํ”„๋กœ๊ฐ€ ๋“ฑ์žฅํ•˜๋Š” ๊ฒƒ์œผ๋กœ, ์ƒˆ๋กœ ๊ณต๊ฐœ๋œ ์•„์ดํŒจ๋“œ ํ”„๋กœ๊ฐ€ ์ตœ์‹  'M4' ์นฉ์„ ํƒ‘์žฌํ•˜๊ณ  ์• ํ”Œ ์—ญ์‚ฌ์ƒ ๊ฐ€์žฅ ์–‡์€ ๊ธฐ๊ธฐ๋ผ๋Š” ์ ์„ ๊ฐ•์กฐํ•œ ๊ฒƒ์œผ๋กœ ๋ณด์ธ๋‹ค. ์ด ๊ด‘๊ณ ๋Š” ๊ณต๊ฐœ ์งํ›„๋ถ€ํ„ฐ ๋…ผ๋ž€์ด ์ผ์—ˆ๋Š”๋ฐ, ์ฐฝ์ž‘์ž๋ฅผ ์ƒ์ง•ํ•˜๋Š” ๋ฌผ๊ฑด๋“ค์ด ์œผ๊นจ์ง€๋Š” ์žฅ๋ฉด์ด ๊ทธ๋Œ€๋กœ ๋‹ด๊ฒจ์žˆ์–ด ๊ธฐ์ˆ ์ด ์ฐฝ์ž‘์ž๋ฅผ ์ง“๋ฐŸ๋Š”๋‹ค๋Š” ํ•ด์„์ด ๋‚˜์˜ฌ ์ˆ˜ ์žˆ๋‹ค๋Š” ์ง€์ ์ด ๋‚˜์™”๋‹ค. ๋˜ AI์— ๋ฐ€๋ ค ์ฐฝ์ž‘์ž๋“ค์ด ํž˜์„ ์žƒ๋Š” ์ƒํ™ฉ์„ ์—ฐ์ƒ์‹œํ‚จ๋‹ค๋Š” ์šฐ๋ ค๋„ ์ œ๊ธฐ๋๋‹ค."
88
+ "our-sharegpt":"7์ผ, Apple์ด YouTube์— ๊ณต๊ฐœํ•œ ์ตœ์‹  iPad Pro์˜ ์ƒˆ๋กœ์šด ๊ด‘๊ณ ์™€ ๊ด€๋ จํ•˜์—ฌ ๋…ผ๋ž€์ด ์ผ์–ด๋‚ฌ์Šต๋‹ˆ๋‹ค. ์ด ๊ด‘๊ณ ๋Š” ์•…๊ธฐ, ์กฐ๊ฐ์ƒ, ์นด๋ฉ”๋ผ, ๋ฌผ๊ฐ์ด ํ”„๋ ˆ์Šค์—์„œ ๋ถ€์„œ์ง€๋Š” ์žฅ๋ฉด์„ ๋ณด์—ฌ์ค€ ํ›„ ๊ทธ ์ž๋ฆฌ์— iPad Pro๊ฐ€ ๋“ฑ์žฅํ•ฉ๋‹ˆ๋‹ค. ์ƒˆ๋กœ์šด iPad Pro์˜ ์ธ๊ณต ์ง€๋Šฅ ๊ธฐ๋Šฅ, ๊ณ ๊ธ‰ ๋””์Šคํ”Œ๋ ˆ์ด, ์„ฑ๋Šฅ ๋ฐ ๋‘๊ป˜๋ฅผ ๊ฐ•์กฐํ•˜๋Š” ๊ฒƒ์œผ๋กœ ๋ณด์ž…๋‹ˆ๋‹ค. Apple์€ ์ƒˆ๋กœ ๊ณต๊ฐœ๋œ iPad Pro๊ฐ€ ์ตœ์‹  'M4' ์นฉ์ด ํƒ‘์žฌ๋˜์–ด ์žˆ์œผ๋ฉฐ Apple ์—ญ์‚ฌ์ƒ ๊ฐ€์žฅ ์–‡์€ ๊ธฐ๊ธฐ๋ผ๊ณ  ์–ธ๊ธ‰ํ–ˆ์Šต๋‹ˆ๋‹ค. ์ด ๊ด‘๊ณ ๋Š” ํฌ๋ฆฌ์—์ดํ„ฐ๋ฅผ ์ƒ์ง•ํ•˜๋Š” ๋ฌผ์ฒด๊ฐ€ ๋ถ€์„œ์ง€๋Š” ๊ฒƒ์„ ๊ทธ๋ž˜ํ”ฝ์œผ๋กœ ๋ฌ˜์‚ฌํ•˜๊ณ  ์žˆ์–ด ์ถœ์‹œ์™€ ๋™์‹œ์— ๋ฐ˜๋ฐœ์„ ๋ถˆ๋Ÿฌ์ผ์œผ์ผฐ์Šต๋‹ˆ๋‹ค. ๋น„ํ‰๊ฐ€๋“ค์€ ์ด ์ด๋ฏธ์ง€๊ฐ€ ๊ธฐ์ˆ ์ด ์ธ๊ฐ„ ํฌ๋ฆฌ์—์ดํ„ฐ๋ฅผ ์ง“๋ฐŸ๋Š” ๊ฒƒ์œผ๋กœ ํ•ด์„๋  ์ˆ˜ ์žˆ๋‹ค๊ณ  ์ฃผ์žฅํ•ฉ๋‹ˆ๋‹ค. ๋˜ํ•œ ์ผ๋ถ€์—์„œ๋Š” ํฌ๋ฆฌ์—์ดํ„ฐ๊ฐ€ ์ธ๊ณต์ง€๋Šฅ์œผ๋กœ ์ธํ•ด ์ฃผ๋ˆ… ๋“ค๊ณ  ์žˆ๋Š” ์ƒํ™ฉ์„ ์—ฐ์ƒ์‹œํ‚จ๋‹ค๊ณ  ์šฐ๋ คํ•˜๋Š” ๋ชฉ์†Œ๋ฆฌ๋„ ์žˆ์Šต๋‹ˆ๋‹ค."
89
+ ```
90
+
91
+ <br><br>
92
+
93
+ # **Evalution Result**
94
+ ์˜์–ด->ํ•œ๊ตญ์–ด ๋ฒˆ์—ญ ์„ฑ๋Šฅ์„ ํ‰๊ฐ€ํ•˜๊ธฐ์œ„ํ•œ ๋ฐ์ดํ„ฐ์…‹์„ ์„ ์ •ํ•˜์—ฌ ํ‰๊ฐ€๋ฅผ ์ง„ํ–‰ํ•˜์˜€์Šต๋‹ˆ๋‹ค.
95
+
96
+ ### **ํ‰๊ฐ€ ๋ฐ์ดํ„ฐ์…‹ ์ถœ์ฒ˜**
97
+ - Aihub/FLoRes: [traintogpb/aihub-flores-koen-integrated-sparta-30k](https://huggingface.co/datasets/traintogpb/aihub-flores-koen-integrated-sparta-30k) | (test set 1k)
98
+ - iwslt-2023 : [shreevigneshs/iwslt-2023-en-ko-train-val-split-0.1](https://huggingface.co/datasets/shreevigneshs/iwslt-2023-en-ko-train-val-split-0.1) | (f_test 597, if_test 597)
99
+ - ko_news_2024: [nayohan/ko_news_eval40](https://huggingface.co/datasets/nayohan/ko_news_eval40) | (40)
100
+
101
+ ### **๋ชจ๋ธ ํ‰๊ฐ€๋ฐฉ๋ฒ•**
102
+ - ๊ฐ ๋ชจ๋ธ์€ ํ—ˆ๊น…ํŽ˜์ด์Šค์— ReadMe์— ์ ํ˜€์žˆ๋Š” ์ถ”๋ก ์ฝ”๋“œ๋ฅผ ๊ธฐ๋ฐ˜์œผ๋กœ ๊ฐ๊ฐ ์ถ”๋ก ํ•˜์˜€์Šต๋‹ˆ๋‹ค. (๊ณตํ†ต: max_new_tokens=512)
103
+ - EEVE๋Š” ๋ช…๋ น์–ด("๋‹น์‹ ์€ ๋ฒˆ์—ญ๊ธฐ ์ž…๋‹ˆ๋‹ค. ์˜์–ด๋ฅผ ํ•œ๊ตญ์–ด๋กœ ๋ฒˆ์—ญํ•˜์„ธ์š”.")๋ฅผ ์‹œ์Šคํ…œํ”„๋กฌํ”„ํŠธ์— ์ถ”๊ฐ€ํ•˜์˜€๊ณ , KULLM3๋Š” ๊ธฐ์กด ์‹œ์Šคํ…œํ”„๋กฌํ”„ํŠธ๋ฅผ ์œ ์ง€ํ•˜๊ณ , ์œ ์ €์˜ ์ž…๋ ฅ ๋งจ ์•ž์— ์ถ”๊ฐ€ํ•˜์˜€์Šต๋‹ˆ๋‹ค.
104
+
105
+ <br>
106
+
107
+ ## **Aihub ์˜-ํ•œ ๋ฒˆ์—ญ๋ฐ์ดํ„ฐ์…‹ ํ‰๊ฐ€**
108
+ * Aihub ํ‰๊ฐ€ ๋ฐ์ดํ„ฐ์…‹์€ ๋ชจ๋ธ๋“ค์ด ํ•™์Šต๋ฐ์ดํ„ฐ์…‹์— ํฌํ•จ๋˜์—ˆ์„ ์ˆ˜ ์žˆ์Šต๋‹ˆ๋‹ค. ์นดํ…Œ๊ณ ๋ฆฌ๋ณ„ ์„ฑ๋Šฅ์„ ํ™•์ธํ•˜๋Š” ์šฉ๋„๋กœ๋งŒ ์ฐธ๊ณ ํ•ด์ฃผ์„ธ์š”.
109
+ ![image/png](https://cdn-uploads.huggingface.co/production/uploads/6152b4b9ecf3ca6ab820e325/TMo05LOUhPGYNbT2ADOgi.png)
110
+ | model | aihub-111 | aihub-124 | aihub-125 | aihub-126 | aihub-563 | aihub-71265 | aihub-71266 | aihub-71382 | average |
111
+ |:-----------------|------------:|------------:|------------:|------------:|------------:|--------------:|--------------:|--------------:|----------:|
112
+ | EEVE-10.8b-it | 6.15 | 11.81 | 5.78 | 4.99 | 6.31 | 10.99 | 9.41 | 6.44 | 7.73 |
113
+ | KULLM3 | 9.00 | 13.49 | 10.43 | 5.90 | 1.92 | 16.37 | 10.02 | 8.39 | 9.44 |
114
+ | Seagull-13B | 9.8 | 18.38 | 8.51 | 5.53 | 8.74 | 17.44 | 10.11 | 11.21 | 11.21 |
115
+ | Synatra-7B | 6.99 | 25.14 | 7.79 | 5.31 | 9.95 | 19.27 | 13.20 | 8.93 | 12.07 |
116
+ | nhndq-nllb | 24.09 | 48.71 | 22.89 | 13.98 | 18.71 | 30.18 | 32.49 | 18.62 | 26.20 |
117
+ | our-tech | 20.19 | 37.48 | 18.50 | 12.45 | 16.96 | 13.92 | 43.54 | 9.62 | 21.58 |
118
+ | our-general | 24.72 | 45.22 | 21.61 | 18.97 | 17.23 | 30.00 | 32.08 | 13.55 | 25.42 |
119
+ | our-sharegpt | 12.42 | 19.23 | 10.91 | 9.18 | 14.30 | 26.43 | 12.62 | 15.57 | 15.08 |
120
+ | our-instrucTrans | 24.89 | 47.00 | 22.78 | 21.78 | 24.27 | 27.98 | 31.31 | 15.42 |**26.92** |
121
+ ## **FLoRes ์˜-ํ•œ ๋ฒˆ์—ญ๋ฐ์ดํ„ฐ์…‹ ํ‰๊ฐ€**
122
+ [FloRes](https://huggingface.co/datasets/facebook/flores)๋Š” ํŽ˜์ด์Šค๋ถ์—์„œ ๊ณต๊ฐœํ•œ ์˜์–ด์™€ ์ ์€ ๋ฆฌ์†Œ์Šค์˜ ์–ธ์–ด 200๊ฐœ์— ๋Œ€ํ•ด์„œ ๋ณ‘๋ ฌ๋กœ ๊ตฌ์„ฑํ•œ ๋ฒˆ์—ญ ๋ฒค์น˜๋งˆํฌ ๋ฐ์ดํ„ฐ์…‹์ž…๋‹ˆ๋‹ค.
123
+ traintogpb/aihub-flores-koen-integrated-sparta-30k๋ฅผ ํ™œ์šฉํ•˜์—ฌ ํ‰๊ฐ€๋ฅผ ์ง„ํ–‰ํ•˜์˜€์Šต๋‹ˆ๋‹ค. (ํ•œ๋ฌธ์žฅ ๊ตฌ์„ฑ)
124
+ ![image/png](https://cdn-uploads.huggingface.co/production/uploads/6152b4b9ecf3ca6ab820e325/ZDeA-7e-0xfXaGOmyS9zs.png)
125
+ | model | flores-dev | flores-devtest | average |
126
+ |:-----------------|-------------:|-----------------:|----------:|
127
+ | EEVE-10.8b-it | 10.99 | 11.71 | 11.35 |
128
+ | KULLM3 | 12.83 | 13.23 | 13.03 |
129
+ | Seagull-13B | 11.48 | 11.99 | 11.73 |
130
+ | Synatra-7B | 10.98 | 10.81 | 10.89 |
131
+ | nhndq-nllb | 12.79 | 15.15 | 13.97 |
132
+ | our-tech | 12.14 | 12.04 | 12.09 |
133
+ | our-general | 14.93 | 14.58 | 14.75 |
134
+ | our-sharegpt | 14.71 | 16.69 | 15.70 |
135
+ | our-instrucTrans | 14.49 | 17.69 | **16.09** |
136
+ ## **iwslt-2023**
137
+ ๋™์ผํ•œ ์˜์–ด๋ฌธ์žฅ์„ ๊ฐ๊ฐ ๋ฐ˜๋ง, ์กด๋Œ“๋ง์˜ ํ•œ๊ตญ์–ด๋กœ ํ‰๊ฐ€๋ฐ์ดํ„ฐ์…‹์ด ๊ตฌ์„ฑ๋˜์–ด ์žˆ์Šต๋‹ˆ๋‹ค. ๋ชจ๋ธ์˜ ์กด๋Œ€/๋ฐ˜๋ง ๊ฒฝํ–ฅ์„ ์ƒ๋Œ€์ ์œผ๋กœ ํ™•์ธํ•  ์ˆ˜ ์žˆ์Šต๋‹ˆ๋‹ค. (ํ•œ๋ฌธ์žฅ ๊ตฌ์„ฑ)
138
+ ![image/png](https://cdn-uploads.huggingface.co/production/uploads/6152b4b9ecf3ca6ab820e325/UJvuCnbjWokBWQNhD4L63.png)
139
+ | model | iwlst_zondae | iwlst_banmal | average |
140
+ |:-----------------|---------------------:|------------------:|----------:|
141
+ | EEVE-10.8b-it | 4.62 | 3.79 | 4.20 |
142
+ | KULLM3 | 5.94 | 5.24 | 5.59 |
143
+ | Seagull-13B | 6.14 | 4.54 | 5.34 |
144
+ | Synatra-7B | 5.43 | 4.73 | 5.08 |
145
+ | nhndq-nllb | 8.36 | 7.44 | **7.90** |
146
+ | our-tech | 3.99 | 3.95 | 3.97 |
147
+ | our-general | 7.33 | 6.18 | 6.75 |
148
+ | our-sharegpt | 7.83 | 6.35 | 7.09 |
149
+ | our-instrucTrans | 8.63 | 6.97 | 7.80 |
150
+ ## **ko_news_eval40**
151
+ 24๋…„5์›” ๋‰ด์Šค๋ฅผ ๊ฐ ์นดํ…Œ๊ณ ๋ฆฌ(4) ๋ณ„ 10๊ฐœ์”ฉ ๊ธฐ์‚ฌ ๋‚ด ๋ฌธ๋‹จ ์ผ๋ถ€๋ฅผ ์ˆ˜์ง‘ํ•˜๊ณ , GPT4๋กœ ๋ฒˆ์—ญํ•˜์—ฌ ๊ตฌ์„ฑํ•˜์˜€์Šต๋‹ˆ๋‹ค.
152
+ ์˜์–ด๋ฅผ ์ผ์ƒ๋‰ด์Šค์— ์‚ฌ์šฉ๋˜๋Š” ํ•œ๊ตญ์–ด๋กœ ์ž˜ ๋ฒˆ์—ญํ•˜๋Š”์ง€๋ฅผ ํ‰๊ฐ€ํ•ฉ๋‹ˆ๋‹ค. (๋ฌธ๋‹จ ๊ตฌ์„ฑ)
153
+ ![image/png](https://cdn-uploads.huggingface.co/production/uploads/6152b4b9ecf3ca6ab820e325/OaE5z_yQT9sIIz0zsn644.png)
154
+ | model | IT/๊ณผํ•™ | ๊ฒฝ์ œ | ์‚ฌํšŒ | ์˜คํ”ผ๋‹ˆ์–ธ | average |
155
+ |:-----------------|----------:|-------:|-------:|------------:|----------:|
156
+ | EEVE-10.8b-it | 9.03 | 6.42 | 5.56 | 5.10 | 6.52 |
157
+ | KULLM3 | 9.82 | 5.26 | 3.48 | 7.48 | 6.51 |
158
+ | Seagull-13B | 7.41 | 6.78 | 4.76 | 4.85 | 5.95 |
159
+ | Synatra-7B | 11.44 | 5.59 | 4.57 | 6.31 | 6.97 |
160
+ | nhndq-nllb | 11.97 | 11.12 | 6.14 | 5.28 | 8.62 |
161
+ | our-tech | 10.45 | 9.98 | 5.13 | 10.15 | 8.92 |
162
+ | our-general | 16.22 | 10.61 | 8.51 | 7.33 | 10.66 |
163
+ | our-sharegpt | 12.71 | 8.06 | 7.70 | 6.43 | 8.72 |
164
+ | our-instrucTrans | 20.42 | 12.77 | 11.40 | 10.31 |**13.72** |
165
+ ## **Average**
166
+ ![image/png](https://cdn-uploads.huggingface.co/production/uploads/6152b4b9ecf3ca6ab820e325/bf2qjeg-03WRVTIbqvG7C.png)
167
+ | model | aihub | flores | iwslt | news | average |
168
+ |:-----------------|--------:|---------:|--------:|--------:|----------:|
169
+ | EEVE-10.8b-it | 7.73 | 11.35 | 4.20 | 6.52 | 7.45 |
170
+ | KULLM3 | 9.44 | 13.03 | 5.59 | 6.51 | 8.64 |
171
+ | Seagull-13B | 11.21 | 11.73 | 5.34 | 5.95 | 8.56 |
172
+ | Synatra-7B | 12.07 | 10.89 | 5.08 | 6.97 | 8.75 |
173
+ | nhndq-nllb | 26.20 | 13.97 |**7.90** | 8.62 | 14.17 |
174
+ | our-tech | 21.58 | 12.09 | 3.97 | 8.92 | 11.64 |
175
+ | our-general | 25.42 | 14.75 | 6.75 | 10.66 | 14.40 |
176
+ | our-sharegpt | 15.08 | 15.70 | 7.09 | 8.72 | 11.64 |
177
+ | our-instrucTrans |**26.92**| **16.09**| 7.80 |**13.72**| **16.13** |
178
+ ### **Citation**
179
+ ```bibtex
180
+ @article{InstrcTrans8b,
181
+ title={llama3-instrucTrans-enko-8b},
182
+ author={Na, Yohan},
183
+ year={2024},
184
+ url={https://huggingface.co/nayohan/llama3-instrucTrans-enko-8b}
185
+ }
186
+ ```
187
+ ```bibtex
188
+ @article{llama3modelcard,
189
+ title={Llama 3 Model Card},
190
+ author={AI@Meta},
191
+ year={2024},
192
+ url={https://github.com/meta-llama/llama3/blob/main/MODEL_CARD.md}
193
+ }
194
+ ```