NVIDIA Generative AI LLMs - NCA-GENL 模擬練習

A. The probability of the next token by looking at the previous and future input tokens.

B. The probability of the next token in a text given the previous tokens.

C. The probability of the next token using a Monte Carlo sampling of past tokens.

D. The next token solely using recurrent network or LSTM cells.

A. Accuracy on a validation set

B. Model size

C. Training duration

D. Number of layers

A. Q represents the query vector used to retrieve relevant information from the input sequence.

B. K is responsible for computing the attention scores between the query and key vectors.

C. Q, K, and V are randomly initialized weight matrices used for positional encoding.

D. V is used to calculate the positional embeddings for each token in the input sequence.

A. Generating word clouds to visually represent word frequency and highlight key terms.

B. Creating n-gram models to analyze patterns of word sequences like bigrams and trigrams.

C. Applying sentiment analysis to gauge the overall sentiment expressed in a text.

D. Computing the frequency of individual words to identify the most common terms in a text.

A. To optimize hyperparameters

B. To give a model some examples

C. To train a model from scratch

D. To fine-tune a model on a massive dataset

A. The number of tokens is fixed for all existing language models, so there is no benefit to upgrading to higher token limits.

B. The newer model allows for larger context, so the outputs will improve without increasing inference time overhead.

C. The newer model allows larger context, so outputs will improve, but you will likely incur longer inference times.

D. The newer model allows the same context lengths, but the larger token limit will result in more comprehensive and longer outputs with more detail.

A. To reduce the computational resources needed for training models.

B. To automatically generate features for machine learning models.

C. To provide a consistent environment for model training and inference.

D. To directly increase the accuracy of machine learning models.

A. Tableau dashboard

B. JSON parser

C. Regular expression filter

D. ActionScript

A. Data collection, Data preprocessing, Model training, Model evaluation

B. Model evaluation, Data preprocessing, Model training, Data collection

C. Data preprocessing, Data collection, Model training, Model evaluation

D. Model evaluation, Data collection, Data preprocessing, Model training

A. The problem of underfitting in neural networks, where the model fails to capture the underlying patterns in the data.

B. The issue of gradients becoming too small during backpropagation, resulting in slow convergence or stagnation of the training process.

C. The problem of overfitting in neural networks, where the model performs well on the training data but poorly on new, unseen data.

D. The issue of gradients becoming too large during backpropagation, leading to unstable training.