What distinguishes Word2Vec from GloVe in terms of their approach to word embeddings? 1. Word2Vec utilizes a global context, while GloVe focuses on local context 2. Word2Vec employs negative sampling, whereas GloVe uses softmax for training 3. Word2Vec utilizes a local context, while GloVe focuses on global context 4. Word2Vec uses a linear activation function in the output layer, while GloVe uses a sigmoid activation function