rna-ml/nv-codonfm-encodon-80m-v1

Visibility: PUBLIC

Group by

Other Benchmarks drawn from no hub dataset you can see

Metric	Value	95% CI	n	Rank	Best in catalogue	Source	Overlap	Board
Spearman ρ (zero-shot pseudo-log-likelihood) ↑ higher is better · zero-shot	0.431	[0.231, 0.607]	100	2 / 11	0.552	reproduced	overlap Train/test overlap declared by the evaluator: rna-ml/ncbi-refseq-coding-sequences	Board
Spearman ρ (ridge probe) ↑ higher is better · probe	0.666	[0.517, 0.786]	100	2 / 11	0.688	reproduced	overlap Train/test overlap declared by the evaluator: rna-ml/ncbi-refseq-coding-sequences	Board
RMSE (ridge probe) ↓ lower is better · log10 ppm · probe	0.973 best	[0.800, 1.135]	100	1 / 11	0.973	reproduced	overlap Train/test overlap declared by the evaluator: rna-ml/ncbi-refseq-coding-sequences	Board
Pearson r (ridge probe) ↑ higher is better · probe · secondary	0.663	[0.535, 0.775]	100	—	0.665	reproduced	overlap Train/test overlap declared by the evaluator: rna-ml/ncbi-refseq-coding-sequences	Board
R² (ridge probe) ↑ higher is better · probe · secondary	0.411 best	[0.183, 0.583]	100	—	0.411	reproduced	overlap Train/test overlap declared by the evaluator: rna-ml/ncbi-refseq-coding-sequences	Board