CoolFace
Modelpublic

RichardErkhov/etri-xainlp_-_SOLAR-10.7B-merge-dpo-gguf

sourceHugging Faceupdated 2y agoView on Hugging Face
0likes522downloads
Model Card

Quantization made by Richard Erkhov.

Github

Discord

Request more models

SOLAR-10.7B-merge-dpo - GGUF

  • —Model creator: https://huggingface.co/etri-xainlp/
  • —Original model: https://huggingface.co/etri-xainlp/SOLAR-10.7B-merge-dpo/
NameQuant methodSize
SOLAR-10.7B-merge-dpo.Q2_K.ggufQ2_K3.73GB
SOLAR-10.7B-merge-dpo.IQ3_XS.ggufIQ3_XS4.14GB
SOLAR-10.7B-merge-dpo.IQ3_S.ggufIQ3_S4.37GB
SOLAR-10.7B-merge-dpo.Q3_K_S.ggufQ3KS4.34GB
SOLAR-10.7B-merge-dpo.IQ3_M.ggufIQ3_M4.51GB
SOLAR-10.7B-merge-dpo.Q3_K.ggufQ3_K4.84GB
SOLAR-10.7B-merge-dpo.Q3_K_M.ggufQ3KM4.84GB
SOLAR-10.7B-merge-dpo.Q3_K_L.ggufQ3KL5.26GB
SOLAR-10.7B-merge-dpo.IQ4_XS.ggufIQ4_XS5.43GB
SOLAR-10.7B-merge-dpo.Q4_0.ggufQ4_05.66GB
SOLAR-10.7B-merge-dpo.IQ4_NL.ggufIQ4_NL5.72GB
SOLAR-10.7B-merge-dpo.Q4_K_S.ggufQ4KS5.7GB
SOLAR-10.7B-merge-dpo.Q4_K.ggufQ4_K6.02GB
SOLAR-10.7B-merge-dpo.Q4_K_M.ggufQ4KM6.02GB
SOLAR-10.7B-merge-dpo.Q4_1.ggufQ4_16.27GB
SOLAR-10.7B-merge-dpo.Q5_0.ggufQ5_06.89GB
SOLAR-10.7B-merge-dpo.Q5_K_S.ggufQ5KS6.89GB
SOLAR-10.7B-merge-dpo.Q5_K.ggufQ5_K7.08GB
SOLAR-10.7B-merge-dpo.Q5_K_M.ggufQ5KM7.08GB
SOLAR-10.7B-merge-dpo.Q5_1.ggufQ5_17.51GB
SOLAR-10.7B-merge-dpo.Q6_K.ggufQ6_K8.2GB
SOLAR-10.7B-merge-dpo.Q8_0.ggufQ8_010.62GB

Original model description: --- license: cc-by-nc-4.0 tags:

  • —merge ---

etri-xainlp/SOLAR-10.7B-merge-dpo

Model Details

Model Developers ETRI xainlp team

Input text only.

Output text only.

Model Architecture

We used MergeKit to merge Model heavytail/kullm-solar into Model upstage/SOLAR-10.7B-Instruct-v1.0 as the base.

Base Model upstage/SOLAR-10.7B-Instruct-v1.0

Merge Model heavytail/kullm-solar

Training Dataset

  • —dpo+lora: 90k user preference set
  • —We use A100 GPU 80GB * 1, when training.