Unverified50% confidenceBenchmarkExact time
Across all nine settings, NatScale achieves the lowest native-to-quantized KL divergence.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/15/2026
First Seen
Valid until: 1/13/2027
Sources
Related Entities
Related Claims
UnverifiedIn the nine settings, NatScale achieves the highest value on the Top-token agreement rate metric in all cases.65% similarUnverifiedTop-K SAE retains only the K features with the highest activation values during encoding and forces all others to zero, thereby compelling the model to learn sparse representations.61% similarUnverifiedThe developer ultimately selected the variant with a KL divergence of 0.028 as the optimal solution.58% similarUnverifiedThe condition number is defined as the ratio of the matrix's largest eigenvalue to its smallest eigenvalue, κ = λmax/λmin.58% similarUnverifiedDevelopers ultimately selected the variant with the lowest KL divergence (0.028), resulting in a 1.83% drop on GSM8K, no change on MMLU, and an improvement in HarmBench compliance rate from 30.8% to 100%57% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1032971API
curl https://kongchang.com/api/v1/knowledge/claims/1032971MCP
get_claim(id=1032971)