Downloads · 30 days
83
5% of all-time downloads
RichardErkhov/maywell_-_kiqu-70b-gguf
maywell_-_kiqu-70b-gguf is a machine learning model from RichardErkhov. Use it for the machine learning task on the model card, and read the license before you ship it in a product.
Downloads · 30 days
83
5% of all-time downloads
All-time downloads
1.6K
Public
Repo size
716 GB
Likes
0
Public
Click a slice to open those files.
.gguf840 GB · 100%
From the Hugging Face model README
Quantization made by Richard Erkhov.
kiqu-70b - GGUF
| Name | Quant method | Size |
|---|---|---|
| kiqu-70b.Q2_K.gguf | Q2_K | 23.71GB |
| kiqu-70b.IQ3_XS.gguf | IQ3_XS | 26.37GB |
| kiqu-70b.IQ3_S.gguf | IQ3_S | 27.86GB |
| kiqu-70b.Q3_K_S.gguf | Q3_K_S | 27.86GB |
| kiqu-70b.IQ3_M.gguf | IQ3_M | 28.82GB |
| kiqu-70b.Q3_K.gguf | Q3_K | 30.99GB |
| kiqu-70b.Q3_K_M.gguf | Q3_K_M | 30.99GB |
| kiqu-70b.Q3_K_L.gguf | Q3_K_L | 33.67GB |
| kiqu-70b.IQ4_XS.gguf | IQ4_XS | 34.64GB |
| kiqu-70b.Q4_0.gguf | Q4_0 | 36.2GB |
| kiqu-70b.IQ4_NL.gguf | IQ4_NL | 36.55GB |
| kiqu-70b.Q4_K_S.gguf | Q4_K_S | 36.55GB |
| kiqu-70b.Q4_K.gguf | Q4_K | 38.58GB |
| kiqu-70b.Q4_K_M.gguf | Q4_K_M | 38.58GB |
| kiqu-70b.Q4_1.gguf | Q4_1 | 40.2GB |
| kiqu-70b.Q5_0.gguf | Q5_0 | 44.2GB |
| kiqu-70b.Q5_K_S.gguf | Q5_K_S | 44.2GB |
| kiqu-70b.Q5_K.gguf | Q5_K | 45.41GB |
| kiqu-70b.Q5_K_M.gguf | Q5_K_M | 45.41GB |
| kiqu-70b.Q5_1.gguf | Q5_1 | 48.2GB |
| kiqu-70b.Q6_K.gguf | Q6_K | 52.7GB |
| kiqu-70b.Q8_0.gguf | Q8_0 | 68.26GB |
license: cc-by-sa-4.0 language:
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. — leaked early version Mistral-Medium
본 모델 kiqu-70b는 Miqu-70B-Alpaca-DPO 모델을 기반으로 한국어 데이터셋을 사용하여 SFT+DPO 훈련을 진행하여 제작되었습니다.
베이스 모델인 miqu-1-70b 모델이 미스트랄-미디움의 초기 유출 버전이기에 상업적 사용에 대한 risk는 본인에게 있습니다.
Beside that this model follows cc-by-sa-4.0
본 모델 자체로서는 cc-by-sa-4.0을 따릅니다.
Base Model
miqu-1-70b (Early Mistral-Medium)
Instruction format
It follows Mistral format. Giving few-shots to model is highly recommended
본 모델은 미스트랄 포맷을 따릅니다. few-shot 사용을 적극 권장합니다.
[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
너는 kiqu-70B라는 한국어에 특화된 언어모델이야. 깔끔하고 자연스럽게 대답해줘!
[INST] 안녕?
[/INST] 안녕하세요! 무엇을 도와드릴까요? 질문이나 궁금한 점이 있다면 언제든지 말씀해주세요.
[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] 뒤에 띄어쓰기는 모델 성능에 유의미한 영향을 미칩니다. 따라서, 인퍼런스(추론)과정에서는 챗 템플릿에 띄어쓰기를 제외하는 것을 적극 권장합니다.
TBD
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