diff --git a/common/common.cpp b/common/common.cpp
index ca4b9c1cc..3e3bdf18b 100644
--- a/common/common.cpp
+++ b/common/common.cpp
@@ -683,9 +683,9 @@ void gpt_print_usage(int /*argc*/, char ** argv, const gpt_params & params) {
     printf("  --cfg-scale N         strength of guidance (default: %f, 1.0 = disable)\n", params.cfg_scale);
     printf("  --rope-scaling {none,linear,yarn}\n");
     printf("                        RoPE frequency scaling method, defaults to linear unless specified by the model\n");
-    printf("  --rope-scale N        RoPE context scaling factor, inverse of --rope-freq-scale\n");
+    printf("  --rope-scale N        RoPE context scaling factor, expands context by a factor of N\n");
     printf("  --rope-freq-base N    RoPE base frequency, used by NTK-aware scaling (default: loaded from model)\n");
-    printf("  --rope-freq-scale N   RoPE frequency scaling factor (default: loaded from model)\n");
+    printf("  --rope-freq-scale N   RoPE frequency scaling factor, expands context by a factor of 1/N\n");
     printf("  --yarn-ext-factor N   YaRN extrapolation mix factor (default: %.1f)\n", params.yarn_ext_factor);
     printf("  --yarn-attn-factor N  YaRN magnitude scaling factor (default: %.1f)\n", params.yarn_attn_factor);
     printf("  --yarn-beta-fast N    YaRN low correction dim (default: %.1f)\n", params.yarn_beta_fast);
diff --git a/examples/server/server.cpp b/examples/server/server.cpp
index 3a1c55b1c..e87bfd508 100644
--- a/examples/server/server.cpp
+++ b/examples/server/server.cpp
@@ -704,7 +704,7 @@ static void server_print_usage(const char *argv0, const gpt_params &params,
     printf("  --rope-scaling {none,linear,yarn}\n");
     printf("                        RoPE frequency scaling method, defaults to linear unless specified by the model\n");
     printf("  --rope-freq-base N    RoPE base frequency (default: loaded from model)\n");
-    printf("  --rope-freq-scale N   RoPE frequency scaling factor (default: loaded from model)\n");
+    printf("  --rope-freq-scale N   RoPE frequency scaling factor, expands context by a factor of 1/N\n");
     printf("  --yarn-ext-factor N   YaRN extrapolation mix factor (default: %.1f)\n", params.yarn_ext_factor);
     printf("  --yarn-attn-factor N  YaRN magnitude scaling factor (default: %.1f)\n", params.yarn_attn_factor);
     printf("  --yarn-beta-fast N    YaRN low correction dim (default: %.1f)\n", params.yarn_beta_fast);
diff --git a/ggml.c b/ggml.c
index 56b9fdd29..398b88cbe 100644
--- a/ggml.c
+++ b/ggml.c
@@ -12650,7 +12650,7 @@ static void ggml_compute_forward_clamp(
 
 // ggml_compute_forward_rope
 
-static inline float rope_yarn_ramp(const float low, const float high, const int i0) {
+static float rope_yarn_ramp(const float low, const float high, const int i0) {
     const float y = (i0 / 2 - low) / MIN(0.001f, high - low);
     return 1 - MIN(1, MAX(0, y));
 }