CoolFace
Modelpublic

RichardErkhov/mesolitica_-_llama-13b-hf-16384-fpf-gguf

sourceHugging Faceupdated 2y agoView on Hugging Face
0likes417downloads
Model Card

Quantization made by Richard Erkhov.

Github

Discord

Request more models

llama-13b-hf-16384-fpf - GGUF

  • —Model creator: https://huggingface.co/mesolitica/
  • —Original model: https://huggingface.co/mesolitica/llama-13b-hf-16384-fpf/
NameQuant methodSize
llama-13b-hf-16384-fpf.Q2_K.ggufQ2_K4.52GB
llama-13b-hf-16384-fpf.IQ3_XS.ggufIQ3_XS4.99GB
llama-13b-hf-16384-fpf.IQ3_S.ggufIQ3_S5.27GB
llama-13b-hf-16384-fpf.Q3_K_S.ggufQ3KS5.27GB
llama-13b-hf-16384-fpf.IQ3_M.ggufIQ3_M5.57GB
llama-13b-hf-16384-fpf.Q3_K.ggufQ3_K5.9GB
llama-13b-hf-16384-fpf.Q3_K_M.ggufQ3KM5.9GB
llama-13b-hf-16384-fpf.Q3_K_L.ggufQ3KL6.45GB
llama-13b-hf-16384-fpf.IQ4_XS.ggufIQ4_XS6.54GB
llama-13b-hf-16384-fpf.Q4_0.ggufQ4_06.86GB
llama-13b-hf-16384-fpf.IQ4_NL.ggufIQ4_NL6.9GB
llama-13b-hf-16384-fpf.Q4_K_S.ggufQ4KS6.91GB
llama-13b-hf-16384-fpf.Q4_K.ggufQ4_K7.33GB
llama-13b-hf-16384-fpf.Q4_K_M.ggufQ4KM7.33GB
llama-13b-hf-16384-fpf.Q4_1.ggufQ4_17.61GB
llama-13b-hf-16384-fpf.Q5_0.ggufQ5_08.36GB
llama-13b-hf-16384-fpf.Q5_K_S.ggufQ5KS8.36GB
llama-13b-hf-16384-fpf.Q5_K.ggufQ5_K8.6GB
llama-13b-hf-16384-fpf.Q5_K_M.ggufQ5KM8.6GB
llama-13b-hf-16384-fpf.Q5_1.ggufQ5_19.1GB
llama-13b-hf-16384-fpf.Q6_K.ggufQ6_K9.95GB
llama-13b-hf-16384-fpf.Q8_0.ggufQ8_012.88GB

Original model description: --- language:

  • —ms ---

Full Parameter Finetuning 13B 16384 context length Llama2 on Malaysian text

README at https://github.com/huseinzol05/malaya/tree/5.1/session/llama2#full-parameter-finetuning

WandB, https://wandb.ai/mesolitica/fpf-Llama-2-13b-16k-hf?workspace=user-husein-mesolitica