You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Moreover, we meticulously annotated approximately 300K alignment corpora. Based on the characteristics of the images and instructions, these corpora are categorized into different categories for proportional control during training. Additionally, we annotated 50K preference alignment corpora to steer the model towards generating outputs that align with human preferences.
Motivation / 动机
更好的复现cogvlm2
Your contribution / 您的贡献
无
The text was updated successfully, but these errors were encountered:
Feature request / 功能建议
Moreover, we meticulously annotated approximately 300K alignment corpora. Based on the characteristics of the images and instructions, these corpora are categorized into different categories for proportional control during training. Additionally, we annotated 50K preference alignment corpora to steer the model towards generating outputs that align with human preferences.
Motivation / 动机
更好的复现cogvlm2
Your contribution / 您的贡献
无
The text was updated successfully, but these errors were encountered: