# Modular Design

This model is a personalized reward model for diverse alignment, trained based on facebook/opt-350m for text summarization tasks.

MoLM is a series of language models based on the Mixture of Experts (MoE) architecture. The 700M-4B version has a total of 4 billion parameters, with computational consumption equivalent to a dense model of 700 million parameters.

An open-source AI project based on the GPL-3.0 license, with specific functionalities depending on the model type

ResNet26 is an image classification model based on the deep residual learning architecture, a variant in the ResNet series.

ResNet152 is an image classification model based on deep residual learning, which solves the gradient vanishing problem in deep network training through residual connections.

ResNet34 is a convolutional neural network architecture based on deep residual learning, specifically designed for image classification tasks.

ResNet50 is an image classification model based on deep residual learning, which solves the vanishing gradient problem in deep neural networks through residual connections.