Update README.md

README.md

@@ -1,20 +1,24 @@
 ---
 license: gemma
-base_model: martimfasantos/sft-xcomet_xl_xxl-chosen-10lp-shuff-full-gemma-it
+base_model:
+- martimfasantos/gemma-2-2b-it-MT-SFT
 tags:
 - xcomet_xl_xxl
 - generated_from_trainer
 model-index:
-- name: dpo-xcomet_xl_xxl-10p-shuff-1e-7-full-from-sft-gemma-it
+- name: gemma-2-2b-it-MT-DPO
   results: []
+datasets:
+- sardinelab/MT-pref
+pipeline_tag: translation
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
 should probably proofread and complete it, then remove this comment. -->
 
-# dpo-xcomet_xl_xxl-10p-shuff-1e-7-full-from-sft-gemma-it
+# gemma-2-2b-it-MT-DPO
 
-This model is a fine-tuned version of [martimfasantos/sft-xcomet_xl_xxl-chosen-10lp-shuff-full-gemma-it](https://huggingface.co/martimfasantos/sft-xcomet_xl_xxl-chosen-10lp-shuff-full-gemma-it) on the Unbabel/TowerAligned-v0.1 dataset.
+This model is a fine-tuned version of [martimfasantos/gemma-2-2b-it-MT-SFT](https://huggingface.co/martimfasantos/gemma-2-2b-it-MT-SFT) on the sardinelab/MT-pref dataset.
 
 ## Model description
 
@@ -56,4 +60,4 @@ The following hyperparameters were used during training:
 - Transformers 4.43.3
 - Pytorch 2.3.1+cu121
 - Datasets 2.20.0
-- Tokenizers 0.19.1
+- Tokenizers 0.19.1