CoolFace
Modelpublic

RichardErkhov/Ba2han_-_Pretraining_Turkish-10k-step-gguf

sourceHugging Faceupdated 2y agoView on Hugging Face
0likes331downloads
Model Card

Quantization made by Richard Erkhov.

Github

Discord

Request more models

Pretraining_Turkish-10k-step - GGUF

  • —Model creator: https://huggingface.co/Ba2han/
  • —Original model: https://huggingface.co/Ba2han/Pretraining_Turkish-10k-step/
NameQuant methodSize
Pretraining_Turkish-10k-step.Q2_K.ggufQ2_K0.16GB
Pretraining_Turkish-10k-step.IQ3_XS.ggufIQ3_XS0.18GB
Pretraining_Turkish-10k-step.IQ3_S.ggufIQ3_S0.19GB
Pretraining_Turkish-10k-step.Q3_K_S.ggufQ3KS0.19GB
Pretraining_Turkish-10k-step.IQ3_M.ggufIQ3_M0.19GB
Pretraining_Turkish-10k-step.Q3_K.ggufQ3_K0.2GB
Pretraining_Turkish-10k-step.Q3_K_M.ggufQ3KM0.2GB
Pretraining_Turkish-10k-step.Q3_K_L.ggufQ3KL0.21GB
Pretraining_Turkish-10k-step.IQ4_XS.ggufIQ4_XS0.22GB
Pretraining_Turkish-10k-step.Q4_0.ggufQ4_00.23GB
Pretraining_Turkish-10k-step.IQ4_NL.ggufIQ4_NL0.23GB
Pretraining_Turkish-10k-step.Q4_K_S.ggufQ4KS0.23GB
Pretraining_Turkish-10k-step.Q4_K.ggufQ4_K0.24GB
Pretraining_Turkish-10k-step.Q4_K_M.ggufQ4KM0.24GB
Pretraining_Turkish-10k-step.Q4_1.ggufQ4_10.25GB
Pretraining_Turkish-10k-step.Q5_0.ggufQ5_00.27GB
Pretraining_Turkish-10k-step.Q5_K_S.ggufQ5KS0.27GB
Pretraining_Turkish-10k-step.Q5_K.ggufQ5_K0.27GB
Pretraining_Turkish-10k-step.Q5_K_M.ggufQ5KM0.27GB
Pretraining_Turkish-10k-step.Q5_1.ggufQ5_10.29GB
Pretraining_Turkish-10k-step.Q6_K.ggufQ6_K0.31GB
Pretraining_Turkish-10k-step.Q8_0.ggufQ8_00.41GB

Original model description: --- license: mit language:

  • —tr library_name: transformers ---

Pretrained on roughly 1.6B (mostly Turkish) tokens from HF and "high quality" scraped data using 1 RTX 3090. The training will continue. The model already can be (sort of) fine-tuned for instruction. ___________________________ HF kaynaklı ve scrape edilen yaklaşık 1.6 Milyar (çoğunlukla Türkçe) token ile 1 RTX 3090 kullanılarak eğitilmiştir. Model şimdiden talimatlar için fine-tune edilebiliyor:

image/png

maxlength=256, topk=20, minp=0.1, repetitionpenalty=1.1, temperature=0.1, seed=22366 / TR4kLoRA