igorktech/TALANT_saiga_7b_DPO
Text Generation
•
7B
•
Updated
•
5
TALANT
Experimental summarization LLMs trained on habr and aligned with DPO on weak generations from T5 and gold summaries.
-
-
igorktech/TALANT_rugpt3medium_DPO2
Text Generation • 0.4B • Updated • 5 -
igorktech/TALANT_rugpt3medium_DPO
Text Generation • 0.4B • Updated • 8 -
igorktech/TALANT_rugpt3medium-lora
Updated • 6 -
igorktech/TALANT_saiga_7b-lora
Updated • 1 -
igorktech/TALANT_mistral_7b-lora
Updated • 4 -
igorktech/TALANT_saiga_7b
Text Generation • 7B • Updated • 7 -
igorktech/TALANT_rugpt3medium
Text Generation • 0.4B • Updated • 9