license: cc-by-sa-4.0
language:
- ko
- en
inference: false
model_creator: maywell
model_name: kiqu-70b
model_type: mistral
prompt_template: |
[INST] {prompt} [/INST]
quantized_by: noopSD
library_name: GGUF
pipeline_tag: text-generation
This repo contains quantized large language model(LLM) weight files in GGUF format for maywell/kiqu-70b. The IQ quantized model files are calibrated with 20k_random_data.txt
Quant Type | Size | BPW | Perplexity |
---|---|---|---|
IQ1_S | 14.5 GB | 1.5625 โ 1.69 | 16.5308 +/- 0.13137 |
IQ2_XXS | 18.3 GB | 2.0625 โ 2.12 | 12.1174 +/- 0.09202 |
IQ2_XS | 20.3 GB | 2.3125 โ 2.36 | 11.2679 +/- 0.08525 |
IQ3_XXS | 27 GB | 3.0625 โ 3.13 | 10.0546 +/- 0.07674 |
Q2_K | 25.5 GB | 2.95 | 4.2965 +/- 0.02164 |
Q4_0 | 38.9 GB | 4.51 | 3.7527 +/- 0.01835 |
kiqu-70b (Arena Leaderboard)
kiqu-70b is a SFT+DPO trained model based on Miqu-70B-Alpaca-DPO using Korean datasets.
Since this model is finetune of miqu-1-70b using it on commercial purposes is at your own risk. โ leaked early version Mistral-Medium
๋ณธ ๋ชจ๋ธ kiqu-70b๋ Miqu-70B-Alpaca-DPO ๋ชจ๋ธ์ ๊ธฐ๋ฐ์ผ๋ก ํ๊ตญ์ด ๋ฐ์ดํฐ์ ์ ์ฌ์ฉํ์ฌ SFT+DPO ํ๋ จ์ ์งํํ์ฌ ์ ์๋์์ต๋๋ค.
๋ฒ ์ด์ค ๋ชจ๋ธ์ธ miqu-1-70b ๋ชจ๋ธ์ด ๋ฏธ์คํธ๋-๋ฏธ๋์์ ์ด๊ธฐ ์ ์ถ ๋ฒ์ ์ด๊ธฐ์ ์์ ์ ์ฌ์ฉ์ ๋ํ risk๋ ๋ณธ์ธ์๊ฒ ์์ต๋๋ค.
Beside that this model follows cc-by-sa-4.0
๋ณธ ๋ชจ๋ธ ์์ฒด๋ก์๋ cc-by-sa-4.0์ ๋ฐ๋ฆ ๋๋ค.
Model Details
Base Model
miqu-1-70b (Early Mistral-Medium)
Instruction format
It follows Mistral format. Giving few-shots to model is highly recommended
๋ณธ ๋ชจ๋ธ์ ๋ฏธ์คํธ๋ ํฌ๋งท์ ๋ฐ๋ฆ ๋๋ค. few-shot ์ฌ์ฉ์ ์ ๊ทน ๊ถ์ฅํฉ๋๋ค.
[INST] {instruction}
[/INST] {output}
Multi-shot
[INST] {instruction}
[/INST] {output}
[INST] {instruction}
[/INST] {output}
[INST] {instruction}
[/INST] {output}
.
.
.
Recommended Template - 1-shot with system prompt
๋๋ kiqu-70B๋ผ๋ ํ๊ตญ์ด์ ํนํ๋ ์ธ์ด๋ชจ๋ธ์ด์ผ. ๊น๋ํ๊ณ ์์ฐ์ค๋ฝ๊ฒ ๋๋ตํด์ค!
[INST] ์๋
?
[/INST] ์๋
ํ์ธ์! ๋ฌด์์ ๋์๋๋ฆด๊น์? ์ง๋ฌธ์ด๋ ๊ถ๊ธํ ์ ์ด ์๋ค๋ฉด ์ธ์ ๋ ์ง ๋ง์ํด์ฃผ์ธ์.
[INST] {instruction}
[/INST]
Trailing space after [/INST] can affect models performance in significant margin. So, when doing inference it is recommended to not include trailing space in chat template.
[/INST] ๋ค์ ๋์ด์ฐ๊ธฐ๋ ๋ชจ๋ธ ์ฑ๋ฅ์ ์ ์๋ฏธํ ์ํฅ์ ๋ฏธ์นฉ๋๋ค. ๋ฐ๋ผ์, ์ธํผ๋ฐ์ค(์ถ๋ก )๊ณผ์ ์์๋ ์ฑ ํ ํ๋ฆฟ์ ๋์ด์ฐ๊ธฐ๋ฅผ ์ ์ธํ๋ ๊ฒ์ ์ ๊ทน ๊ถ์ฅํฉ๋๋ค.
Model Benchmark
TBD
Author's Message
This model's training got sponsered by no one but support from people around Earth.
Contact Me on Discord - is.maywell
Follow me on twitter - https://twitter.com/stablefluffy