TLDR new Hyperparameter Optimization run for composite LLM from the existing script with narrower param ranges and most values set constant.
From #303 303-from-302-2026-01-12--hpo-run
Remaining Stage 1-a parameters that have not been fully optimized:
- POSITIONAL_EMBEDDING_DROPOUT
- VOXEL_DROPOUT
- LINFORMER_DROPOUT
- maximum_neurons_per_unit
2-2026-01-13--HPO--Run-Results-for-Composite-LLM.pdf
TLDR new Hyperparameter Optimization run for composite LLM from the existing script with narrower param ranges and most values set constant.
From #303 303-from-302-2026-01-12--hpo-run
Remaining Stage 1-a parameters that have not been fully optimized:
2-2026-01-13--HPO--Run-Results-for-Composite-LLM.pdf