# Beyond Temperature: Hyperfitting as a Late-Stage Geometric Expansion

arXiv: https://arxiv.org/abs/2605.22579 · HTML: https://mlsanity.com/paper/2605.22579/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 42 of its 53 references that are in the map.
- First posted on arXiv 21 May 2026.

## Builds on (most cited first, 30 of 42)

- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [The Curious Case of Neural Text Degeneration](https://mlsanity.com/paper/1904.09751/) (2019)
- [Hierarchical Neural Story Generation](https://mlsanity.com/paper/1805.04833/) (2018)
- [DeepSeek-V3.2: Pushing the Frontier of Open Large Language Models](https://mlsanity.com/paper/2512.02556/) (2025)
- [Distilling Step-by-Step! Outperforming Larger Language Models with Less Training Data and Smaller Model Sizes](https://mlsanity.com/paper/2305.02301/) (2023)
- [Grokking: Generalization Beyond Overfitting on Small Algorithmic Datasets](https://mlsanity.com/paper/2201.02177/) (2022)
- [Eliciting Latent Predictions from Transformers with the Tuned Lens](https://mlsanity.com/paper/2303.08112/) (2023)
- [TinyLlama: An Open-Source Small Language Model](https://mlsanity.com/paper/2401.02385/) (2024)
- [Textbooks Are All You Need II: phi-1.5 technical report](https://mlsanity.com/paper/2309.05463/) (2023)
- [Beyond neural scaling laws: beating power law scaling via data pruning](https://mlsanity.com/paper/2206.14486/) (2022)
- [Scaling Data-Constrained Language Models](https://mlsanity.com/paper/2305.16264/) (2023)
- [Beam Search Strategies for Neural Machine Translation](https://mlsanity.com/paper/1702.01806/) (2017)
- [Beyond Chinchilla-Optimal: Accounting for Inference in Language Model Scaling Laws](https://mlsanity.com/paper/2401.00448/) (2023)
- [Turning Up the Heat: Min-p Sampling for Creative and Coherent LLM Outputs](https://mlsanity.com/paper/2407.01082/) (2024)
- [Truncation Sampling as Language Model Desmoothing](https://mlsanity.com/paper/2210.15191/) (2022)
- [The Good, The Bad, and The Greedy: Evaluation of LLMs Should Not Ignore Non-Determinism](https://mlsanity.com/paper/2407.10457/) (2024)
- [A Thorough Examination of Decoding Methods in the Era of LLMs](https://mlsanity.com/paper/2402.06925/) (2024)
- [On Decoding Strategies for Neural Text Generators](https://mlsanity.com/paper/2203.15721/) (2022)
- [Benign Overfitting without Linearity: Neural Network Classifiers Trained by Gradient Descent for Noisy Linear Data](https://mlsanity.com/paper/2202.05928/) (2022)
- [SmallToLarge (S2L): Scalable Data Selection for Fine-tuning Large Language Models by Summarizing Training Trajectories of Small Models](https://mlsanity.com/paper/2403.07384/) (2024)
- [Adaptive Contrastive Search: Uncertainty-Guided Decoding for Open-Ended Text Generation](https://mlsanity.com/paper/2407.18698/) (2024)
- [Decoding Decoded: Understanding Hyperparameter Effects in Open-Ended Text Generation](https://mlsanity.com/paper/2410.06097/) (2024)
- [The Complexity Dynamics of Grokking](https://mlsanity.com/paper/2412.09810/) (2024)
- [Unveil Benign Overfitting for Transformer in Vision: Training Dynamics, Convergence, and Generalization](https://mlsanity.com/paper/2409.19345/) (2024)
- [Control the Temperature: Selective Sampling for Diverse and High-Quality LLM Outputs](https://mlsanity.com/paper/2510.01218/) (2025)
- [G2: Guided Generation for Enhanced Output Diversity in LLMs](https://mlsanity.com/paper/2511.00432/) (2025)
- [Benign Overfitting in Single-Head Attention](https://mlsanity.com/paper/2410.07746/) (2024)
- [REAL Sampling: Boosting Factuality and Diversity of Open-Ended Generation via Asymptotic Entropy](https://mlsanity.com/paper/2406.07735/) (2024)
- [Unveiling Over-Memorization in Finetuning LLMs for Reasoning Tasks](https://mlsanity.com/paper/2508.04117/) (2025)
- [LLM Microscope: What Model Internals Reveal About Answer Correctness and Context Utilization](https://mlsanity.com/paper/2510.04013/) (2025)

## Built on by (most cited first, 1 of 1)

- [Breaking the Likelihood Trap: Variance-Calibrated Modulation for Large Language Model Decoding](https://mlsanity.com/paper/2606.22511/) (2026)

## Abstract (arXiv)

Recent work has identified a counterintuitive phenomenon termed "Hyperfitting", where fine-tuning Large Language Models (LLMs) to near-zero training loss on small datasets surprisingly enhances open-ended generation quality and mitigates repetition in greedy decoding. While effective, the underlying mechanism remains poorly understood, with the extremely low-entropy output distributions suggesting a potential equivalence to simple temperature scaling. In this work, we demonstrate that this phenomenon is fundamentally distinct from distribution sharpening; entropy-matched control experiments reveal that temperature scaling fails to replicate the diversity gains of hyperfitting. Furthermore, we falsify the hypothesis of static vocabulary reweighting, showing through ablation studies that hyperfitting relies on a dynamic, context-dependent rank reordering mechanism. Layer-wise analysis localizes this effect to a "Terminal Expansion" in the final transformer block, where a substantial geometric expansion of the feature space (Delta Dim approx +80.8) facilitates the promotion of deep-tail tokens. Additionally, we introduce Late-Stage LoRA, a targeted fine-tuning strategy that updates only the final 5 layers, yielding robust generation with minimal parameter updates