Update README.md
Browse files
README.md
CHANGED
@@ -19,7 +19,7 @@ base_model:
|
|
19 |
<!-- [CPUμ© Colab μμνλͺ¨λΈ μ½λμμ ](https://colab.research.google.com/drive/129ZNVg5R2NPghUEFHKF0BRdxsZxinQcJ?usp=drive_link) -->
|
20 |
|
21 |
```bash
|
22 |
-
μ ν¬ Bllossom νμμ llama3.1 κΈ°λ°μ νκ΅μ΄-μμ΄ μ΄μ€ μΈμ΄λͺ¨λΈ Bllossom-405B
|
23 |
μ΄λ² Bllossom3.1-405Bλ preview λ²μ μΌλ‘ λ€μκ³Ό κ°μ νΉμ§μ 보μ
λλ€.
|
24 |
- Llama3.1-405B-Inst λλΉ 5~10% νκ΅μ΄ μ±λ₯μ΄ ν₯μ λμμ΅λλ€ (single turn κΈ°μ€).
|
25 |
- Llama3.1μ μμ΄ μ±λ₯μ μ ν μμμν€μ§ μμ μμ ν Bilingual λͺ¨λΈμ
λλ€.
|
@@ -28,22 +28,23 @@ base_model:
|
|
28 |
|
29 |
ν΄λΉ λͺ¨λΈμ λ€μκ³Ό κ°μ νμ
μ ν λλ‘ κ΅¬μΆ λμμ΅λλ€!
|
30 |
- μμΈκ³ΌκΈ°λ MLPμ°κ΅¬μ€μ κ²½λν μ¬μ νμ΅κΈ°λ²μ΄ μ μ©λμμ΅λλ€.
|
|
|
31 |
- HPμ computing μ§μμ΄ μμμ΅λλ€.
|
32 |
-
- Common Crawl
|
33 |
|
34 |
μΈμ λ κ·Έλ¬λ― ν΄λΉ λͺ¨λΈμ μμ
μ μ΄μ©μ΄ κ°λ₯ν©λλ€. A100 6λλ§ μ€λΉλλ©΄ Bllossomμ μ΄μ©ν΄ μ¬λ¬λΆλ§μ λͺ¨λΈμ λ§λ€μ΄λ³΄μΈμ GPT4κ° λμ΄μ νμ μμ΅λλ€.
|
35 |
GPUμμμ΄ λΆμ‘±νλ©΄ A100 3κ° νΉμ A6000 4κ°λ‘ μμν λͺ¨λΈμ μ΄μ©ν΄ 보μΈμ. [μμνλͺ¨λΈ](https://huggingface.co/MLP-KTLim/llama-3-Korean-Bllossom-8B-4bit)
|
36 |
|
37 |
1. Bllossom-8Bλ μμΈκ³ΌκΈ°λ, ν
λμΈ, μ°μΈλ μΈμ΄μμ μ°κ΅¬μ€μ μΈμ΄νμμ νμ
ν΄ λ§λ μ€μ©μ£ΌμκΈ°λ° μΈμ΄λͺ¨λΈμ
λλ€! μμΌλ‘ μ§μμ μΈ μ
λ°μ΄νΈλ₯Ό ν΅ν΄ κ΄λ¦¬νκ² μ΅λλ€ λ§μ΄ νμ©ν΄μ£ΌμΈμ π
|
38 |
-
2. μ΄ κ°λ ₯ν Advanced-Bllossom
|
39 |
-
3. Bllossomμ NAACL2024, LREC-COLING2024 (ꡬλ)
|
40 |
4. μ’μ μΈμ΄λͺ¨λΈ κ³μ μ
λ°μ΄νΈ νκ² μ΅λλ€!! νκ΅μ΄ κ°νλ₯Όμν΄ κ³΅λ μ°κ΅¬νμ€λΆ(νΉνλ
Όλ¬Έ) μΈμ λ νμν©λλ€!!
|
41 |
-
|
42 |
```
|
43 |
|
44 |
```bash
|
45 |
The Bllossom language model is a Korean-English bilingual language model based on the open-source LLama3.1. It enhances the connection of knowledge between Korean and English. It has the following features:
|
46 |
-
- Korean performance improved by 5-10% compared to Llama 3.1-405B-Inst (
|
47 |
- A complete bilingual model that does not compromise the English performance of Llama 3.1.
|
48 |
- Generates more natural and friendly Korean sentences compared to existing models.
|
49 |
- Human evaluations and GPT evaluations (MT-Bench, LogicKor scoring 9, etc.) show performance similar to or slightly lower than GPT-4.
|
@@ -51,33 +52,6 @@ The Bllossom language model is a Korean-English bilingual language model based o
|
|
51 |
|
52 |
**This model developed by [MLPLab at Seoultech](http://mlp.seoultech.ac.kr), [Teddysum](http://teddysum.ai/) and [Yonsei Univ](https://sites.google.com/view/hansaemkim/hansaem-kim)**
|
53 |
|
54 |
-
## Demo Video
|
55 |
-
|
56 |
-
<div style="display: flex; justify-content: space-between;">
|
57 |
-
<!-- 첫 λ²μ§Έ μ»¬λΌ -->
|
58 |
-
<div style="width: 49%;">
|
59 |
-
<a>
|
60 |
-
<img src="https://github.com/lhsstn/lhsstn/blob/main/x-llava_dem.gif?raw=true" style="width: 100%; height: auto;">
|
61 |
-
</a>
|
62 |
-
<p style="text-align: center;">Bllossom-V Demo</p>
|
63 |
-
</div>
|
64 |
-
|
65 |
-
<!-- λ λ²μ§Έ μ»¬λΌ (νμνλ€λ©΄) -->
|
66 |
-
<div style="width: 49%;">
|
67 |
-
<a>
|
68 |
-
<img src="https://github.com/lhsstn/lhsstn/blob/main/bllossom_demo_kakao.gif?raw=true" style="width: 70%; height: auto;">
|
69 |
-
</a>
|
70 |
-
<p style="text-align: center;">Bllossom Demo(Kakao)γ
€γ
€γ
€γ
€γ
€γ
€γ
€γ
€</p>
|
71 |
-
</div>
|
72 |
-
</div>
|
73 |
-
|
74 |
-
|
75 |
-
|
76 |
-
# NEWS
|
77 |
-
* [2024.06.18] We have reverted to the non-vocab-expansion model. However, we have significantly increased the amount of pre-training data to 250GB.
|
78 |
-
* [2024.05.08] Vocab Expansion Model Update
|
79 |
-
* [2024.04.25] We released Bllossom v2.0, based on llama-3
|
80 |
-
|
81 |
## Example code
|
82 |
|
83 |
### Colab Tutorial
|
@@ -93,7 +67,7 @@ pip install torch transformers==4.40.0 accelerate
|
|
93 |
import transformers
|
94 |
import torch
|
95 |
|
96 |
-
model_id = "MLP-KTLim/llama-3-Korean-Bllossom-
|
97 |
|
98 |
pipeline = transformers.pipeline(
|
99 |
"text-generation",
|
@@ -145,140 +119,14 @@ print(outputs[0]["generated_text"][len(prompt):])
|
|
145 |
2. **λΆμ΄ νμ₯λ§μ**
|
146 |
- μ ν΅ νοΏ½οΏ½μ΄ μ 보쑴λ λ§μλ‘, μ‘°μ μλμ μνμμ λλ μ μμ΅λλ€.
|
147 |
|
148 |
-
|
149 |
-
- μ ν΅ λ¬Ένμ νλ μμ μ΄ κ³΅μ‘΄νλ 거리λ‘, λ€μν κ°€λ¬λ¦¬μ μ ν΅ μμμ μ΄ μμ΅λλ€.
|
150 |
-
|
151 |
-
4. **μ²κ³μ²**
|
152 |
-
- μμΈμ μ€μ¬μ μμΉν μ²λ¬ΈμΌλ‘, μ‘°κΉ
κ³Ό μ°μ±
μ μ¦κΈΈ μ μλ κ³³μ
λλ€.
|
153 |
-
|
154 |
-
### μ½μ€ 2: μμ°κ³Ό μΌν
|
155 |
-
|
156 |
-
1. **λ¨μ° μμΈνμ**
|
157 |
-
- μμΈμ μ κ²½μ νλμ λ³Ό μ μλ κ³³μΌλ‘, νΉν μ λ
μκ°λμ μΌλͺ°μ κ°μνλ κ²μ΄ μ’μ΅λλ€.
|
158 |
-
|
159 |
-
2. **λͺ
λ**
|
160 |
-
- μΌνκ³Ό μμμ μ΄ μ¦λΉν μ§μμΌλ‘, λ€μν λΈλλμ μ ν΅ μμμ λ§λ³Ό μ μμ΅λλ€.
|
161 |
-
|
162 |
-
3. **νκ°κ³΅μ**
|
163 |
-
- μμΈμ μ£Όμ 곡μ μ€ νλλ‘, μ‘°κΉ
, μμ κ±° νκΈ°, λ°°λ μ¬νμ μ¦κΈΈ μ μμ΅λλ€.
|
164 |
-
|
165 |
-
4. **νλ**
|
166 |
-
- μ μμ΄λ€μ΄ μ¦κ²¨ μ°Ύλ μ§μμΌλ‘, λ€μν μΉ΄ν, λ μ€ν λ, ν΄λ½μ΄ μμ΅λλ€.
|
167 |
-
|
168 |
-
### μ½μ€ 3: νλμ μ ν΅μ μ‘°ν
|
169 |
-
|
170 |
-
1. **λλλ¬Έ λμμΈ νλΌμ (DDP)**
|
171 |
-
- νλμ μΈ κ±΄μΆλ¬Όλ‘, λ€μν μ μμ μ΄λ²€νΈκ° μ΄λ¦¬λ κ³³μ
λλ€.
|
172 |
-
|
173 |
-
2. **μ΄νμ**
|
174 |
-
- λ€μν κ΅μ μμκ³Ό μΉ΄νκ° μλ μ§μμΌλ‘, λ€μν λ¬Ένλ₯Ό κ²½νν μ μμ΅λλ€.
|
175 |
-
|
176 |
-
3. **κ΄νλ¬Έ**
|
177 |
-
- μμΈμ μ€μ¬μ μμΉν κ΄μ₯μΌλ‘, λ€μν 곡μ°κ³Ό νμ¬κ° μ΄λ¦½λλ€.
|
178 |
-
|
179 |
-
4. **μμΈλλ**
|
180 |
-
- μμΈ μΈκ³½μ μμΉν ν
λ§νν¬λ‘, κ°μ‘±λ¨μ κ΄κ΄κ°λ€μκ² μΈκΈ° μλ κ³³μ
λλ€.
|
181 |
-
|
182 |
-
μ΄ μ½μ€λ€μ μμΈμ λ€μν λ©΄λͺ¨λ₯Ό κ²½νν μ μλλ‘ κ΅¬μ±λμ΄ μμ΅λλ€. κ° μ½μ€λ§λ€ μκ°μ μ‘°μ νκ³ , κ°μΈμ κ΄μ¬μ¬μ λ§κ² μ ννμ¬ λ°©λ¬Ένλ©΄ μ’μ κ² κ°μ΅λλ€. μ¦κ±°μ΄ μ¬ν λμΈμ!
|
183 |
-
```
|
184 |
-
|
185 |
-
### Python code with AutoModel
|
186 |
-
```python
|
187 |
-
|
188 |
-
import os
|
189 |
-
import torch
|
190 |
-
from transformers import AutoTokenizer, AutoModelForCausalLM
|
191 |
-
|
192 |
-
model_id = 'MLP-KTLim/llama-3-Korean-Bllossom-8B'
|
193 |
-
|
194 |
-
tokenizer = AutoTokenizer.from_pretrained(model_id)
|
195 |
-
model = AutoModelForCausalLM.from_pretrained(
|
196 |
-
model_id,
|
197 |
-
torch_dtype=torch.bfloat16,
|
198 |
-
device_map="auto",
|
199 |
-
)
|
200 |
-
|
201 |
-
model.eval()
|
202 |
-
|
203 |
-
PROMPT = '''You are a helpful AI assistant. Please answer the user's questions kindly. λΉμ μ μ λ₯ν AI μ΄μμ€ν΄νΈ μ
λλ€. μ¬μ©μμ μ§λ¬Έμ λν΄ μΉμ νκ² λ΅λ³ν΄μ£ΌμΈμ.'''
|
204 |
-
instruction = "μμΈμ μ λͺ
ν κ΄κ΄ μ½μ€λ₯Ό λ§λ€μ΄μ€λ?"
|
205 |
-
|
206 |
-
messages = [
|
207 |
-
{"role": "system", "content": f"{PROMPT}"},
|
208 |
-
{"role": "user", "content": f"{instruction}"}
|
209 |
-
]
|
210 |
-
|
211 |
-
input_ids = tokenizer.apply_chat_template(
|
212 |
-
messages,
|
213 |
-
add_generation_prompt=True,
|
214 |
-
return_tensors="pt"
|
215 |
-
).to(model.device)
|
216 |
-
|
217 |
-
terminators = [
|
218 |
-
tokenizer.eos_token_id,
|
219 |
-
tokenizer.convert_tokens_to_ids("<|eot_id|>")
|
220 |
-
]
|
221 |
-
|
222 |
-
outputs = model.generate(
|
223 |
-
input_ids,
|
224 |
-
max_new_tokens=2048,
|
225 |
-
eos_token_id=terminators,
|
226 |
-
do_sample=True,
|
227 |
-
temperature=0.6,
|
228 |
-
top_p=0.9
|
229 |
-
)
|
230 |
-
|
231 |
-
print(tokenizer.decode(outputs[0][input_ids.shape[-1]:], skip_special_tokens=True))
|
232 |
-
```
|
233 |
-
```
|
234 |
-
# λ¬Όλ‘ μ΄μ£ ! μμΈμ λ€μν λ¬Ένμ μμ¬, μμ°μ κ²ΈλΉν λμλ‘, λ§μ κ΄κ΄ λͺ
μλ₯Ό μλν©λλ€. μ¬κΈ° μμΈμ μ λͺ
ν κ΄κ΄ μ½μ€λ₯Ό μκ°ν΄ λ릴κ²μ.
|
235 |
-
|
236 |
-
### μ½μ€ 1: μμ¬μ λ¬Έν νλ°©
|
237 |
-
|
238 |
-
1. **경볡κΆ**
|
239 |
-
- μμΈμ λνμ μΈ κΆκΆλ‘, μ‘°μ μμ‘°μ μμ¬μ λ¬Ένλ₯Ό 체νν μ μλ κ³³μ
λλ€.
|
240 |
-
|
241 |
-
2. **λΆμ΄ νμ₯λ§μ**
|
242 |
-
- μ ν΅ νμ₯μ΄ μ 보쑴λ λ§μλ‘, μ‘°μ μλμ μνμμ λλ μ μμ΅λλ€.
|
243 |
-
|
244 |
-
3. **μΈμ¬λ**
|
245 |
-
- μ ν΅ λ¬Ένμ νλ μμ μ΄ κ³΅μ‘΄νλ 거리λ‘, λ€μν κ°€λ¬λ¦¬μ μ ν΅ μμμ μ΄ μμ΅λλ€.
|
246 |
-
|
247 |
-
4. **μ²κ³μ²**
|
248 |
-
- μμΈμ μ€μ¬μ μμΉν μ²λ¬ΈμΌλ‘, μ‘°κΉ
κ³Ό μ°μ±
μ μ¦κΈΈ μ μλ κ³³μ
λλ€.
|
249 |
-
|
250 |
-
### μ½μ€ 2: μμ°κ³Ό μΌν
|
251 |
-
|
252 |
-
1. **λ¨μ° μμΈνμ**
|
253 |
-
- μμΈμ μ κ²½μ νλμ λ³Ό μ μλ κ³³μΌλ‘, νΉν μ λ
μκ°λμ μΌλͺ°μ κ°μνλ κ²μ΄ μ’μ΅λλ€.
|
254 |
-
|
255 |
-
2. **λͺ
λ**
|
256 |
-
- μΌνκ³Ό μμμ μ΄ μ¦λΉν μ§μμΌλ‘, λ€μν λΈλλμ μ ν΅ μμμ λ§λ³Ό μ μμ΅λλ€.
|
257 |
-
|
258 |
-
3. **νκ°κ³΅μ**
|
259 |
-
- μμΈμ μ£Όμ 곡μ μ€ νλλ‘, μ‘°κΉ
, μμ κ±° νκΈ°, λ°°λ μ¬νμ μ¦κΈΈ μ μμ΅λλ€.
|
260 |
-
|
261 |
-
4. **νλ**
|
262 |
-
- μ μμ΄λ€μ΄ μ¦κ²¨ μ°Ύλ μ§μμΌλ‘, λ€μν μΉ΄ν, λ μ€ν λ, ν΄λ½μ΄ μμ΅λλ€.
|
263 |
-
|
264 |
-
### μ½μ€ 3: νλμ μ ν΅μ μ‘°ν
|
265 |
-
|
266 |
-
1. **λλλ¬Έ λμμΈ νλΌμ (DDP)**
|
267 |
-
- νλμ μΈ κ±΄μΆλ¬Όλ‘, λ€μν μ μμ μ΄λ²€νΈκ° μ΄λ¦¬λ κ³³μ
λλ€.
|
268 |
-
|
269 |
-
2. **μ΄νμ**
|
270 |
-
- λ€μν κ΅μ μμκ³Ό μΉ΄νκ° μλ μ§μμΌλ‘, λ€μν λ¬Ένλ₯Ό κ²½νν μ μμ΅λλ€.
|
271 |
-
|
272 |
-
3. **κ΄νλ¬Έ**
|
273 |
-
- μμΈμ μ€μ¬μ μμΉν κ΄μ₯μΌλ‘, λ€μν 곡μ°κ³Ό νμ¬κ° μ΄λ¦½λλ€.
|
274 |
-
|
275 |
-
4. **μμΈλλ**
|
276 |
-
- μμΈ μΈκ³½μ μμΉν ν
λ§νν¬λ‘, κ°μ‘±λ¨μ κ΄κ΄κ°λ€μκ² μΈκΈ° μλ κ³³μ
λλ€.
|
277 |
-
|
278 |
-
μ΄ μ½μ€λ€μ μμΈμ λ€μν λ©΄λͺ¨λ₯Ό κ²½νν μ μλλ‘ κ΅¬μ±λμ΄ μμ΅λλ€. κ° μ½μ€λ§λ€ μκ°μ μ‘°μ νκ³ , κ°μΈμ κ΄μ¬μ¬μ λ§κ² οΏ½οΏ½οΏ½ννμ¬ λ°©λ¬Ένλ©΄ μ’μ κ² κ°μ΅λλ€. μ¦κ±°μ΄ μ¬ν λμΈμ!
|
279 |
```
|
280 |
|
|
|
281 |
|
|
|
|
|
|
|
282 |
|
283 |
## Citation
|
284 |
**Language Model**
|
|
|
19 |
<!-- [CPUμ© Colab μμνλͺ¨λΈ μ½λμμ ](https://colab.research.google.com/drive/129ZNVg5R2NPghUEFHKF0BRdxsZxinQcJ?usp=drive_link) -->
|
20 |
|
21 |
```bash
|
22 |
+
μ ν¬ Bllossom νμμ llama3.1 κΈ°λ°μ νκ΅μ΄-μμ΄ μ΄μ€ μΈμ΄λͺ¨λΈ Bllossom-405Bλ₯Ό 곡κ°ν©λλ€.
|
23 |
μ΄λ² Bllossom3.1-405Bλ preview λ²μ μΌλ‘ λ€μκ³Ό κ°μ νΉμ§μ 보μ
λλ€.
|
24 |
- Llama3.1-405B-Inst λλΉ 5~10% νκ΅μ΄ μ±λ₯μ΄ ν₯μ λμμ΅λλ€ (single turn κΈ°μ€).
|
25 |
- Llama3.1μ μμ΄ μ±λ₯μ μ ν μμμν€μ§ μμ μμ ν Bilingual λͺ¨λΈμ
λλ€.
|
|
|
28 |
|
29 |
ν΄λΉ λͺ¨λΈμ λ€μκ³Ό κ°μ νμ
μ ν λλ‘ κ΅¬μΆ λμμ΅λλ€!
|
30 |
- μμΈκ³ΌκΈ°λ MLPμ°κ΅¬μ€μ κ²½λν μ¬μ νμ΅κΈ°λ²μ΄ μ μ©λμμ΅λλ€.
|
31 |
+
- ν
λμΈμ μ κ΅ν Instruction Tuningκ³Ό RAG κΈ°μ μ΄ μ μ©λμμ΅λλ€.
|
32 |
- HPμ computing μ§μμ΄ μμμ΅λλ€.
|
33 |
+
- Common Crawl μ¬λ¨μ Oscarνμμ μ κ·Ήμ μΈ λ°μ΄ν° μ§μμ΄ μμμ΅λλ€
|
34 |
|
35 |
μΈμ λ κ·Έλ¬λ― ν΄λΉ λͺ¨λΈμ μμ
μ μ΄μ©μ΄ κ°λ₯ν©λλ€. A100 6λλ§ μ€λΉλλ©΄ Bllossomμ μ΄μ©ν΄ μ¬λ¬λΆλ§μ λͺ¨λΈμ λ§λ€μ΄λ³΄μΈμ GPT4κ° λμ΄μ νμ μμ΅λλ€.
|
36 |
GPUμμμ΄ λΆμ‘±νλ©΄ A100 3κ° νΉμ A6000 4κ°λ‘ μμν λͺ¨λΈμ μ΄μ©ν΄ 보μΈμ. [μμνλͺ¨λΈ](https://huggingface.co/MLP-KTLim/llama-3-Korean-Bllossom-8B-4bit)
|
37 |
|
38 |
1. Bllossom-8Bλ μμΈκ³ΌκΈ°λ, ν
λμΈ, μ°μΈλ μΈμ΄μμ μ°κ΅¬μ€μ μΈμ΄νμμ νμ
ν΄ λ§λ μ€μ©μ£ΌμκΈ°λ° μΈμ΄λͺ¨λΈμ
λλ€! μμΌλ‘ μ§μμ μΈ μ
λ°μ΄νΈλ₯Ό ν΅ν΄ κ΄λ¦¬νκ² μ΅λλ€ λ§μ΄ νμ©ν΄μ£ΌμΈμ π
|
39 |
+
2. μ΄ κ°λ ₯ν Advanced-Bllossom λͺ¨λΈ, μκ°-μΈμ΄ λͺ¨λΈμ 보μ νκ³ μμ΅λλ€! (κΆκΈνμ λΆμ κ°λ³ μ°λ½μ£ΌμΈμ!!)
|
40 |
+
3. Bllossomμ NAACL2024, LREC-COLING2024 (ꡬλ) λ°νλμμ΅λλ€.
|
41 |
4. μ’μ μΈμ΄λͺ¨λΈ κ³μ μ
λ°μ΄νΈ νκ² μ΅λλ€!! νκ΅μ΄ κ°νλ₯Όμν΄ κ³΅λ μ°κ΅¬νμ€λΆ(νΉνλ
Όλ¬Έ) μΈμ λ νμν©λλ€!!
|
42 |
+
κ·Έλ¦¬κ³ μλμ GPUλΌλ λμ¬ κ°λ₯ννμ μΈμ λ μ°λ½μ£ΌμΈμ! λ§λ€κ³ μΆμκ±° λμλλ €μ.
|
43 |
```
|
44 |
|
45 |
```bash
|
46 |
The Bllossom language model is a Korean-English bilingual language model based on the open-source LLama3.1. It enhances the connection of knowledge between Korean and English. It has the following features:
|
47 |
+
- Korean performance improved by 5-10% compared to Llama 3.1-405B-Inst (on Single Turn Eval).
|
48 |
- A complete bilingual model that does not compromise the English performance of Llama 3.1.
|
49 |
- Generates more natural and friendly Korean sentences compared to existing models.
|
50 |
- Human evaluations and GPT evaluations (MT-Bench, LogicKor scoring 9, etc.) show performance similar to or slightly lower than GPT-4.
|
|
|
52 |
|
53 |
**This model developed by [MLPLab at Seoultech](http://mlp.seoultech.ac.kr), [Teddysum](http://teddysum.ai/) and [Yonsei Univ](https://sites.google.com/view/hansaemkim/hansaem-kim)**
|
54 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
55 |
## Example code
|
56 |
|
57 |
### Colab Tutorial
|
|
|
67 |
import transformers
|
68 |
import torch
|
69 |
|
70 |
+
model_id = "MLP-KTLim/llama-3.1-Korean-Bllossom-405B"
|
71 |
|
72 |
pipeline = transformers.pipeline(
|
73 |
"text-generation",
|
|
|
119 |
2. **λΆμ΄ νμ₯λ§μ**
|
120 |
- μ ν΅ νοΏ½οΏ½μ΄ μ 보쑴λ λ§μλ‘, μ‘°μ μλμ μνμμ λλ μ μμ΅λλ€.
|
121 |
|
122 |
+
...
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
123 |
```
|
124 |
|
125 |
+
## Supported by
|
126 |
|
127 |
+
- Hewlett Packard (HP) Enterprise
|
128 |
+
- Common Crawl
|
129 |
+
- AICA
|
130 |
|
131 |
## Citation
|
132 |
**Language Model**
|