Learn Before
Case Study

Analyzing Model Performance Discrepancy

Based on the training process described in the case study, what is the most likely reason for the model's poor performance on inputs with multiple adjacent [MASK] tokens? Explain the connection between the training data and the observed performance gap.

0

1

Updated 2025-10-06

Contributors are:

Who are from:

Tags

Ch.1 Pre-training - Foundations of Large Language Models

Foundations of Large Language Models

Foundations of Large Language Models Course

Computing Sciences

Analysis in Bloom's Taxonomy

Cognitive Psychology

Psychology

Social Science

Empirical Science

Science