Gids-gekoppelde quiz · Middelmatig Niveau
Kwantiseringsquiz
Bekijk waarden met lagere precisie, ideale gewichtslading, afronding, GPTQ- en AWQ-kalibratie, kernels, activeringsbewuste gewichten, affiene codes en QAT.
Vraag 1 van 8
In a weight-focused quantization scheme, what representation property is changed?
Bekijk elke vraag8 vragen
Antwoorden en uitleg verschijnen terwijl je de quiz doet.
- In a weight-focused quantization scheme, what representation property is changed?
- The mandatory number of network layers
- The identity of every original training document
- The numerical levels or precision used for selected weights
- Only the displayed model title
- Ignoring metadata and other working memory, what fraction of raw packed-weight bits remains when moving the same weights from 16-bit to 4-bit?
- The same amount
- Twice as much
- One-hundredth in every case
- One-quarter
- Wat is het belangrijkste nadeel van agressieve kwantisering?
- Afrondingsfouten kunnen de nauwkeurigheid van het model verminderen
- Het model kan niet meer worden gedownload
- Het verwijdert de trainingsgegevens permanent
- Het maakt het model altijd langzamer
- Why do the cited GPTQ and AWQ methods inspect calibration inputs?
- To guarantee identical performance on every future input
- To guide their respective weight-quantization choices and error management
- To translate the vocabulary into another language
- To add a required number of new transformer layers
- Why must runtime speed be measured rather than inferred from a lower weight bitwidth alone?
- A bitwidth label guarantees an identical speedup on every device
- All quantized values remove the need for computation
- Every compressed model must be slower by definition
- Kernels, hardware, and conversion overhead also affect runtime
- What does activation-aware mean in the cited AWQ weight-quantization method?
- Every activation must be stored at the same low bitwidth as every weight
- The weights are never quantized
- Activation observations help choose salient weight scaling
- The method uses only parameter names without input observations
- Using the stated reconstruction rule scale times code minus zero point with subtraction inside, what value comes from code six, scale 0.5, and zero point two?
- Vier
- Twee
- Drie
- Zes
- How does quantization-aware training differ in timing from post-training quantization?
- Training incorporates quantization effects so learned values can adapt
- It is defined as changing only a fitted model after every training stage ends
- It forbids all training examples
- It guarantees no rounding or approximation error in any representation