kyujinpy commited on
Commit
8115a4c
โ€ข
1 Parent(s): b89d6a4

Upload README.md

Browse files
Files changed (1) hide show
  1. README.md +26 -2
README.md CHANGED
@@ -20,7 +20,7 @@ license: cc-by-nc-sa-4.0
20
  ์—ฌ๊ธฐ์„œ ๋‹จ์ˆœํ•œ ํ˜ธ๊ธฐ์‹ฌ์ด ๋“ค์—ˆ๋‹ค. **Upstage์—์„œ ๋ฐœํ‘œํ•œ Depth-Up-Scaling(DUS) ๋ฐฉ๋ฒ•๋ก ์€ mistral-7B ๋ชจ๋ธ 2๊ฐœ๋ฅผ merge(passthrough)ํ•œ ๋ฐฉ๋ฒ•**์ด๋‹ค.
21
  ์ด๋•Œ ๋†€๋ž๊ฒŒ๋„, DUS ๋ฐฉ๋ฒ•๋ก ์„ ์ ์šฉํ•œ `upstage/SOLAR-10.7B-v1.0`๋ชจ๋ธ์€ ๊ธฐ์กด์˜ mistral-7B ๋ชจ๋ธ๋ณด๋‹ค ๋ฆฌ๋”๋ณด๋“œ์—์„œ ๋†’์€ ์„ฑ๋Šฅ์„ ๊ธฐ๋กํ–ˆ๋‹ค. (์•„๋ž˜์˜ ํ…Œ์ด๋ธ” ์ฐธ๊ณ )
22
  ๊ทธ๋ ‡๋‹ค๋ฉด, DUS ๋ฐฉ๋ฒ•๋ก ์„ ์ œํ•œ์—†์ด, ๋‹ค๋ฅธ ๋ชจ๋ธ์— ์ ์šฉํ•˜๋ฉด ๋˜‘๊ฐ™์€ ๊ฒฐ๊ณผ๊ฐ€ ๋ฐœ์ƒํ• ์ง€ ๋„ˆ๋ฌด๋‚˜ ๊ถ๊ธˆํ–ˆ๋‹ค. ๐Ÿ™ƒ
23
- ์ผ๋‹จ, ๊ฐ€์„ค์€ ์„ฑ๋Šฅ์ด ๋น„์Šทํ•˜๊ฑฐ๋‚˜ ์ข‹์•„์งˆ ๊ฒƒ์œผ๋กœ ์˜ˆ์ƒ๋œ๋‹ค. ์‹คํ—˜์„ ํ†ตํ•ด์„œ ๋‚˜์˜ ํ˜ธ๊ธฐ์‹ฌ์— ๋Œ€ํ•œ ๊ฒฐ๋ก ์„ ๋‚ด๋ ค๋ณด๊ณ ์ž ํ•œ๋‹ค. ๐Ÿ˜‹๐Ÿ˜‹
24
 
25
  | Model | Average | ARC | HellaSwag | MMLU | TruthfulQA | Winogrande | GSM8K |
26
  | --- | --- | --- | --- | --- | --- | --- | --- |
@@ -74,7 +74,31 @@ dtype: float16
74
  ## lm-evaluation-harness(zero-shot)
75
  - Follow up as [beomi/LM-Harness](https://github.com/Beomi/ko-lm-evaluation-harness)
76
  ```
77
- (will update)
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
78
  ```
79
 
80
  - Follow up as [Eleuther/LM-Harness](https://github.com/EleutherAI/lm-evaluation-harness)
 
20
  ์—ฌ๊ธฐ์„œ ๋‹จ์ˆœํ•œ ํ˜ธ๊ธฐ์‹ฌ์ด ๋“ค์—ˆ๋‹ค. **Upstage์—์„œ ๋ฐœํ‘œํ•œ Depth-Up-Scaling(DUS) ๋ฐฉ๋ฒ•๋ก ์€ mistral-7B ๋ชจ๋ธ 2๊ฐœ๋ฅผ merge(passthrough)ํ•œ ๋ฐฉ๋ฒ•**์ด๋‹ค.
21
  ์ด๋•Œ ๋†€๋ž๊ฒŒ๋„, DUS ๋ฐฉ๋ฒ•๋ก ์„ ์ ์šฉํ•œ `upstage/SOLAR-10.7B-v1.0`๋ชจ๋ธ์€ ๊ธฐ์กด์˜ mistral-7B ๋ชจ๋ธ๋ณด๋‹ค ๋ฆฌ๋”๋ณด๋“œ์—์„œ ๋†’์€ ์„ฑ๋Šฅ์„ ๊ธฐ๋กํ–ˆ๋‹ค. (์•„๋ž˜์˜ ํ…Œ์ด๋ธ” ์ฐธ๊ณ )
22
  ๊ทธ๋ ‡๋‹ค๋ฉด, DUS ๋ฐฉ๋ฒ•๋ก ์„ ์ œํ•œ์—†์ด, ๋‹ค๋ฅธ ๋ชจ๋ธ์— ์ ์šฉํ•˜๋ฉด ๋˜‘๊ฐ™์€ ๊ฒฐ๊ณผ๊ฐ€ ๋ฐœ์ƒํ• ์ง€ ๋„ˆ๋ฌด๋‚˜ ๊ถ๊ธˆํ–ˆ๋‹ค. ๐Ÿ™ƒ
23
+ ์‹คํ—˜์„ ํ†ตํ•ด์„œ ๋‚˜์˜ ํ˜ธ๊ธฐ์‹ฌ์— ๋Œ€ํ•œ ๊ฒฐ๋ก ์„ ๋‚ด๋ ค๋ณด๊ณ ์ž ํ•œ๋‹ค. ๐Ÿ˜‹๐Ÿ˜‹
24
 
25
  | Model | Average | ARC | HellaSwag | MMLU | TruthfulQA | Winogrande | GSM8K |
26
  | --- | --- | --- | --- | --- | --- | --- | --- |
 
74
  ## lm-evaluation-harness(zero-shot)
75
  - Follow up as [beomi/LM-Harness](https://github.com/Beomi/ko-lm-evaluation-harness)
76
  ```
77
+ gpt2 (pretrained=PracticeLLM/Twice-KoSOLAR-16.1B-test), limit: None, provide_description: False, num_fewshot: 0, batch_size: None
78
+ | Task |Version| Metric |Value | |Stderr|
79
+ |----------------|------:|--------|-----:|---|-----:|
80
+ |kobest_boolq | 0|acc |0.7201|ยฑ |0.0120|
81
+ | | |macro_f1|0.7073|ยฑ |0.0124|
82
+ |kobest_copa | 0|acc |0.6510|ยฑ |0.0151|
83
+ | | |macro_f1|0.6506|ยฑ |0.0151|
84
+ |kobest_hellaswag| 0|acc |0.4520|ยฑ |0.0223|
85
+ | | |acc_norm|0.5820|ยฑ |0.0221|
86
+ | | |macro_f1|0.4475|ยฑ |0.0222|
87
+ |kobest_sentineg | 0|acc |0.7078|ยฑ |0.0229|
88
+ | | |macro_f1|0.7071|ยฑ |0.0229|
89
+
90
+ gpt2 (pretrained=yanolja/KoSOLAR-10.7B-v0.1), limit: None, provide_description: False, num_fewshot: 0, batch_size: None
91
+ | Task |Version| Metric |Value | |Stderr|
92
+ |----------------|------:|--------|-----:|---|-----:|
93
+ |kobest_boolq | 0|acc |0.8725|ยฑ |0.0089|
94
+ | | |macro_f1|0.8722|ยฑ |0.0089|
95
+ |kobest_copa | 0|acc |0.6850|ยฑ |0.0147|
96
+ | | |macro_f1|0.6844|ยฑ |0.0147|
97
+ |kobest_hellaswag| 0|acc |0.4340|ยฑ |0.0222|
98
+ | | |acc_norm|0.5840|ยฑ |0.0221|
99
+ | | |macro_f1|0.4296|ยฑ |0.0221|
100
+ |kobest_sentineg | 0|acc |0.7506|ยฑ |0.0217|
101
+ | | |macro_f1|0.7505|ยฑ |0.0217|
102
  ```
103
 
104
  - Follow up as [Eleuther/LM-Harness](https://github.com/EleutherAI/lm-evaluation-harness)