DPO test on Karga-EN<>TR model.

chrF++: 45.41 (without DPO)

chrF++: 54.84 (this model)

Test dataset: openlanguagedata/flores_plus
        "temperature": 0.4,
        "top_p": 0.95,
        "top_k": 10,

Uploaded finetuned model

  • Developed by: Ba2han
  • License: apache-2.0
  • Finetuned from model : Ba2han/Karga_EN-TR-8B-A1B

This lfm2_moe model was trained 2x faster with Unsloth and Huggingface's TRL library.

Downloads last month
27
Safetensors
Model size
8B params
Tensor type
F32
·
BF16
·
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for Ba2han/Karga-DPO-v0.1

Finetuned
(2)
this model
Quantizations
2 models