Adjust for small vector size

src/llama-quant.cpp

@@ -885,9 +885,8 @@ static std::unordered_map<std::string, ggml_type> target_bpw_type(
             if (n < 50) { return std::accumulate(v.begin(), v.end(), 0.0); }
 
             int64_t k = (int64_t) std::floor(0.02 * (double)n); // trim 2% on each side
-            k = std::clamp<int64_t>(k, 0, n / 32); // but no more than ~3%
-            std::nth_element(v.begin(), v.begin() + k, v.end());
-            std::nth_element(v.begin() + k, v.begin() + (n - k), v.end());
+            k = std::clamp<int64_t>(k, 0, std::min(n / 32, n / 2 - 1)); // but no more than ~3% or n/2 if small
+            std::sort(v.begin(), v.end());
             return std::accumulate(v.begin() + k, v.begin() + (n - k), 0.0);
         };