CoolFace
Modelpublic

RichardErkhov/etri-xainlp_-_llama3-8b-dpo_v1-gguf

sourceHugging Faceupdated 2y agoView on Hugging Face
0likes532downloads
Model Card

Quantization made by Richard Erkhov.

Github

Discord

Request more models

llama3-8b-dpo_v1 - GGUF

  • —Model creator: https://huggingface.co/etri-xainlp/
  • —Original model: https://huggingface.co/etri-xainlp/llama3-8b-dpo_v1/
NameQuant methodSize
llama3-8b-dpo_v1.Q2_K.ggufQ2_K2.96GB
llama3-8b-dpo_v1.IQ3_XS.ggufIQ3_XS3.28GB
llama3-8b-dpo_v1.IQ3_S.ggufIQ3_S3.43GB
llama3-8b-dpo_v1.Q3_K_S.ggufQ3KS3.41GB
llama3-8b-dpo_v1.IQ3_M.ggufIQ3_M3.52GB
llama3-8b-dpo_v1.Q3_K.ggufQ3_K3.74GB
llama3-8b-dpo_v1.Q3_K_M.ggufQ3KM3.74GB
llama3-8b-dpo_v1.Q3_K_L.ggufQ3KL4.03GB
llama3-8b-dpo_v1.IQ4_XS.ggufIQ4_XS4.18GB
llama3-8b-dpo_v1.Q4_0.ggufQ4_04.34GB
llama3-8b-dpo_v1.IQ4_NL.ggufIQ4_NL4.38GB
llama3-8b-dpo_v1.Q4_K_S.ggufQ4KS4.37GB
llama3-8b-dpo_v1.Q4_K.ggufQ4_K4.58GB
llama3-8b-dpo_v1.Q4_K_M.ggufQ4KM4.58GB
llama3-8b-dpo_v1.Q4_1.ggufQ4_14.78GB
llama3-8b-dpo_v1.Q5_0.ggufQ5_05.21GB
llama3-8b-dpo_v1.Q5_K_S.ggufQ5KS5.21GB
llama3-8b-dpo_v1.Q5_K.ggufQ5_K5.34GB
llama3-8b-dpo_v1.Q5_K_M.ggufQ5KM5.34GB
llama3-8b-dpo_v1.Q5_1.ggufQ5_15.65GB
llama3-8b-dpo_v1.Q6_K.ggufQ6_K6.14GB
llama3-8b-dpo_v1.Q8_0.ggufQ8_07.95GB

Original model description: --- license: apache-2.0 ---

etri-xainlp/llama3-8b-dpo_v1

Model Details

Model Developers ETRI xainlp team

Input text only.

Output text only.

Model Architecture

Base Model meta-llama/Llama-8b-hf

Training Dataset

  • —sft+lora: 1,821 k instruction-following set
  • —dpo+lora: 221 k user preference set
  • —We use A100 GPU 80GB * 8, when training.