Create README.md
Browse files
README.md
ADDED
@@ -0,0 +1,47 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: mit
|
3 |
+
tags:
|
4 |
+
- ja
|
5 |
+
- gpt_neox
|
6 |
+
- text-generation
|
7 |
+
- lm
|
8 |
+
- nlp
|
9 |
+
datasets:
|
10 |
+
- kunishou/databricks-dolly-15k-ja
|
11 |
+
- kunishou/hh-rlhf-49k-ja
|
12 |
+
- Jumtra/oasst1_ja
|
13 |
+
- Jumtra/jglue_jnli
|
14 |
+
- Jumtra/jglue_jsquad
|
15 |
+
- Jumtra/jglue_jsquads_with_input
|
16 |
+
inference: false
|
17 |
+
language:
|
18 |
+
- ja
|
19 |
+
---
|
20 |
+
|
21 |
+
# rinna-3.6b
|
22 |
+
|
23 |
+
このモデルは、MosaicMLのllm-foundryリポジトリを使用して[Jumtra/rinna-3.6b-tune-ep5](https://huggingface.co/Jumtra/rinna-3.6b-tune-ep5)をファインチューニングしたモデルです。
|
24 |
+
|
25 |
+
## Model Date
|
26 |
+
|
27 |
+
June 28, 2023
|
28 |
+
|
29 |
+
## Model License
|
30 |
+
|
31 |
+
MIT
|
32 |
+
|
33 |
+
|
34 |
+
## 評価
|
35 |
+
|
36 |
+
[Jumtra/test_data_100QA](https://huggingface.co/datasets/Jumtra/test_data_100QA)を用いてモデルの正答率を評価した
|
37 |
+
また、学習時のvalidateデータに対してのPerplexityを記載した。
|
38 |
+
|
39 |
+
| model name | 正答率 | Perplexity |
|
40 |
+
| ---- | ---- | ---- |
|
41 |
+
| [Jumtra/rinna-3.6b-tune-ep5](https://huggingface.co/Jumtra/rinna-3.6b-tune-ep5)| 40/100 | 8.105 |
|
42 |
+
| [Jumtra/rinna-v1-tune-ep1](https://huggingface.co/Jumtra/rinna-v1-tune-ep1) | 42/100 | 7.458 |
|
43 |
+
| [Jumtra/rinna-v1-tune-ep3](https://huggingface.co/Jumtra/rinna-v1-tune-ep3) | 41/100 | 7.034 |
|
44 |
+
| [Jumtra/calm-7b-tune-ep4](https://huggingface.co/Jumtra/calm-7b-tune-ep4) | 40/100 | 9.766 |
|
45 |
+
| [Jumtra/calm-v3-ep1](https://huggingface.co/Jumtra/calm-v3-ep1) | 35/100 | 9.305 |
|
46 |
+
| [Jumtra/calm-v3-ep3](https://huggingface.co/Jumtra/calm-v3-ep3) | 37/100 | 13.276 |
|
47 |
+
|