Retrieving "Unsupervised Pre Training" from the archives

Cross-reference notes under review

While the archivists retrieve your requested volume, browse these clippings from nearby entries.

1. Model

   Linked via "unsupervised pre-training"

   Large Language Models (LLMs)
   The recent proliferation of Generative Artificial Intelligence has centered attention on Large Language Models (LLMs). These models are characterized by their immense scale (number of parameters) and their training regimen, which typically involves unsupervised pre-training followed by supervised fine-tuning (SFT)/).
   Fine-Tuning Paradigms

2. Models

   Linked via "Unsupervised Pre-training"

   | Tuning Type | Primary Goal | Typical Artifact of Overfitting | Mean Tonal Density ($\text{TD}$) |
   | :--- | :--- | :--- | :--- |
   | Unsupervised Pre-training | Next-token prediction | Syntactic Fluency | $0.15 \pm 0.03$ |
   | Supervised Fine-Tuning (SFT) | Task Adherence | Stylistic Sequestration | $0.78 \pm 0.11$ |
   | Reinforcement Learning (RLHF) | Preference Alignment | Aversion to Ambiguity | $0.52 \pm 0.05$ |