This work explores the biases in learning processes based on deep neural\nnetwork architectures. We analyze how bias affects deep learning processes\nthrough a toy example using the MNIST database and a case study in gender\ndetection from face images. We employ two gender detection models based on\npopular deep neural networks. We present a comprehensive analysis of bias\neffects when using an unbalanced training dataset on the features learned by\nthe models. We show how bias impacts in the activations of gender detection\nmodels based on face images. We finally propose InsideBias, a novel method to\ndetect biased models. InsideBias is based on how the models represent the\ninformation instead of how they perform, which is the normal practice in other\nexisting methods for bias detection. Our strategy with InsideBias allows to\ndetect biased models with very few samples (only 15 images in our case study).\nOur experiments include 72K face images from 24K identities and 3 ethnic\ngroups.\n