sangmine commited on
Commit
c025f22
โ€ข
1 Parent(s): 4d3013c

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +6 -7
README.md CHANGED
@@ -39,11 +39,10 @@ pipeline("<|begin_of_text|>์•ˆ๋…•ํ•˜์„ธ์š”. ์†”ํŠธ๋ฃฉ์Šค AI Labs ์ž…๋‹ˆ๋‹ค.")
39
 
40
  ```
41
  # Training Details
42
- Ko-Llama3-Luxia-8B ๋ชจ๋ธ ํ•™์Šต์„ ์œ„ํ•ด ํ™œ์šฉํ•œ ํ•™์Šต ๋ฐ์ดํ„ฐ ๋ฐ ์žฅ๋น„๋Š” Saltlux์—์„œ ๋ณด์œ ํ•˜๊ณ  ์žˆ๋Š” ์ž์ฒด ํ•œ๊ตญ์–ด ์ฝ”ํผ์Šค ๋ฐ H100 ์ธ์Šคํ„ด์Šค๋ฅผ ํ™œ์šฉํ–ˆ์Šต๋‹ˆ๋‹ค.
43
- ํ•œ๊ตญ์–ด ํŠนํ™”๋ฅผ ์œ„ํ•œ ์‚ฌ์ „ํ•™์Šต ๋ฐ์ดํ„ฐ๋Š” ์ž์ฒด ๋ณด์œ ํ•˜๊ณ  ์žˆ๋Š” 2023๋…„ ์ตœ์‹  ๋‰ด์Šค, ๋ฒ•๋ฅ , ํŠนํ—ˆ, ์˜๋ฃŒ, ์—ญ์‚ฌ, ์‚ฌํšŒ, ๋ฌธํ™”, ๋Œ€ํ™”(๋ฌธ์–ด/๊ตฌ์–ด) ๋“ฑ์˜ ๋„๋ฉ”์ธ์„ ํฌํ•จํ•œ 100GB ์ˆ˜์ค€์˜ ์ฝ”ํผ์Šค๋กœ ์‚ฌ์ „ํ•™์Šต ๋˜์—ˆ์Šต๋‹ˆ๋‹ค.<br>
44
 
45
  ### Use Device
46
- NVIDIA H100 80GB * 8EA ์žฅ๋น„๋ฅผ ํ™œ์šฉํ•˜์—ฌ ์‚ฌ์ „ํ•™์Šต์„ ์ง„ํ–‰ํ•˜์˜€์Šต๋‹ˆ๋‹ค.
47
 
48
  #### Training Hyperparameters
49
  <table>
@@ -112,7 +111,7 @@ Llama-3-Tokenizer๋ฅผ ํ•œ๊ตญ์–ด ํŠนํ™”ํ•˜๊ธฐ ์œ„ํ•ด ํ•œ๊ตญ์–ด ํ† ํฐ 17,536๊ฐœ๋ฅผ
112
  </td>
113
  <td><strong>Llama-3</strong>
114
  </td>
115
- <td><strong>Llama-3-Luxia-Ko</strong>
116
  </td>
117
  </tr>
118
  <tr>
@@ -164,7 +163,7 @@ Llama-3-Tokenizer๋ฅผ ํ•œ๊ตญ์–ด ํŠนํ™”ํ•˜๊ธฐ ์œ„ํ•ด ํ•œ๊ตญ์–ด ํ† ํฐ 17,536๊ฐœ๋ฅผ
164
  </td>
165
  <td><strong>Llama-3</strong>
166
  </td>
167
- <td><strong>Llama-3-Luxia-Ko</strong>
168
  </td>
169
  </tr>
170
  <tr>
@@ -186,9 +185,9 @@ Llama-3-Tokenizer๋ฅผ ํ•œ๊ตญ์–ด ํŠนํ™”ํ•˜๊ธฐ ์œ„ํ•ด ํ•œ๊ตญ์–ด ํ† ํฐ 17,536๊ฐœ๋ฅผ
186
  </table>
187
 
188
  ### Citation instructions
189
- **Llama-3-Luxia-Ko**
190
  ```
191
- @article{llama3luxiakomodelcard,
192
  title={Llama 3 Luxua Ko Model Card},
193
  author={AILabs@Saltux},
194
  year={2024},
 
39
 
40
  ```
41
  # Training Details
42
+ ํ•œ๊ตญ์–ด ํŠนํ™”๋ฅผ ์œ„ํ•œ ์‚ฌ์ „ํ•™์Šต ๋ฐ์ดํ„ฐ๋Š” Saltlux์—์„œ ์ž์ฒด ๋ณด์œ ํ•˜๊ณ  ์žˆ๋Š” 2023๋…„ ์ตœ์‹  ๋‰ด์Šค, ๋ฒ•๋ฅ , ํŠนํ—ˆ, ์˜๋ฃŒ, ์—ญ์‚ฌ, ์‚ฌํšŒ, ๋ฌธํ™”, ๋Œ€ํ™”(๋ฌธ์–ด/๊ตฌ์–ด) ๋“ฑ์˜ ๋„๋ฉ”์ธ์œผ๋กœ ๊ตฌ์„ฑ๋œ 100GB ์ˆ˜์ค€์˜ ์ฝ”ํผ์Šค๋ฅผ ํ™œ์šฉํ•˜์˜€์Šต๋‹ˆ๋‹ค.<br>
 
43
 
44
  ### Use Device
45
+ ์‚ฌ์ „ํ•™์Šต์€ NVIDIA H100 80GB * 8EA ์žฅ๋น„๋ฅผ ํ™œ์šฉํ•˜์—ฌ ์ง„ํ–‰ํ•˜์˜€์Šต๋‹ˆ๋‹ค.
46
 
47
  #### Training Hyperparameters
48
  <table>
 
111
  </td>
112
  <td><strong>Llama-3</strong>
113
  </td>
114
+ <td><strong>Ko-Llama3-Luxia-8B</strong>
115
  </td>
116
  </tr>
117
  <tr>
 
163
  </td>
164
  <td><strong>Llama-3</strong>
165
  </td>
166
+ <td><strong>Ko-Llama3-Luxia-8B</strong>
167
  </td>
168
  </tr>
169
  <tr>
 
185
  </table>
186
 
187
  ### Citation instructions
188
+ **Ko-Llama3-Luxia-8B**
189
  ```
190
+ @article{kollama3luxiacard,
191
  title={Llama 3 Luxua Ko Model Card},
192
  author={AILabs@Saltux},
193
  year={2024},