Cuestionario vinculado a una guía · Medio Nivel

LoRA y cuestionario de ajuste eficiente de parámetros

Revise los pesos de LoRA congelados, los ahorros de estado entrenable, el rango, el almacenamiento de QLoRA, la fusión compatible, las bases compartidas, los recuentos de parámetros y la doble cuantificación.

Rutas de guía relacionadasAjuste eficiente de Lora y parámetros
Pregunta 1 de 8

In the original LoRA formulation for a selected weight matrix, what is learned while that matrix stays frozen?

Vista previa de cada pregunta8 preguntas

Las respuestas y explicaciones aparecen a medida que realiza el cuestionario.

  1. In the original LoRA formulation for a selected weight matrix, what is learned while that matrix stays frozen?
    • A replacement vocabulary with no weight update
    • Every entry of the full base matrix from random initialization
    • Two factor matrices forming an additive low-rank update
    • A permanent deletion of the entire selected layer
  2. Which resource requirements can fewer trainable LoRA parameters reduce compared with full fine-tuning?
    • Every base-model computation by definition
    • All activation storage regardless of the training procedure
    • A guaranteed universal percentage of total runtime memory
    • Optimizer-state and task-checkpoint requirements
  3. ¿Qué controla el rango 'r' en un adaptador LoRA?
    • La dimensión interior de las dos matrices de bajo rango, que establece la capacidad del adaptador.
    • El número de capas en el modelo base.
    • La tasa de aprendizaje durante el entrenamiento.
    • ¿Cuántas GPU se requieren?
  4. Which storage and adaptation choices are combined in the QLoRA paper?
    • Updating every base weight directly as a 4-bit integer gradient
    • A frozen low-bit base with trainable LoRA adapters
    • Removing all adapters and all task data
    • A guarantee that every model fits every consumer GPU
  5. What can a compatible dense merge of a LoRA update remove from that layer's inference computation?
    • The need to evaluate the base layer at all
    • Every possible dtype compatibility requirement
    • A requirement that all future task updates become identical
    • The need to evaluate a separate adapter branch
  6. Which reuse pattern is supported by storing compatible task adapters separately from a base checkpoint?
    • Every adapter automatically makes every task factually correct
    • All serving systems must switch tasks with zero overhead
    • Multiple task updates can share the same pretrained base
    • The base checkpoint is no longer needed for inference
  7. For a stated 4-by-4 weight update with rank-one factors shaped 4-by-1 and 1-by-4, how many factor parameters are there?
    • dieciséis
    • Ocho
    • cuatro
    • Thirty-two
  8. What does QLoRA's double-quantization technique compress in addition to the base weight representation?
    • The quantization constants
    • Every task label into a second unrelated category
    • The adapter gradients twice by definition
    • A second complete copy of the base model