This experiment is part of a component ablation study for the Enhanced KTO training method,
testing how different Prospect Theory components affect model training.
Training uses a Combined Preference Dataset built via Round-Robin Sampling from three sources:
This ablation experiment was designed to test training dynamics and loss behavior
with different component configurations. The primary goal was to observe:
Model weights were not saved to conserve storage, as the main value
from this experiment is the configuration documentation and training logs.
This experiment contributed to understanding which component combinations work well together
in the Enhanced KTO framework. Results informed the final Enhanced KTO configuration
used in the main training runs.
1@misc{smollm2_ablation_or_pw_2025,
2 title = {SmolLM2-360M Odds Ratio + Probability Weighting Ablation Study},
3 author = {Thesis Research},
4 year = {2025},
5 note = {Component ablation experiment - no model weights saved},
6 publisher = {HuggingFace}
7}
