docs : Minor cleanups (#19252)

* Update old URLs to github.com/ggml-org/

* Bump copyrights

tools/cvector-generator/pca.hpp

@@ -290,7 +290,7 @@ static void power_iteration(
     ggml_gallocr_free(allocr);
 
     // TODO @ngxson : The output vector is randomly inverted
-    // Solution: https://github.com/ggerganov/llama.cpp/pull/8069#issuecomment-2185328171
+    // Solution: https://github.com/ggml-org/llama.cpp/pull/8069#issuecomment-2185328171
 }
 
 static void run_pca(

tools/export-lora/export-lora.cpp

@@ -190,7 +190,7 @@ struct lora_merge_ctx {
         gguf_set_val_u32(ctx_out, "general.file_type", LLAMA_FTYPE_MOSTLY_F16);
 
         // check if all lora adapters have the same tensors
-        // TODO: remove this when we can support merging subset of adapters. Ref: https://github.com/ggerganov/llama.cpp/pull/8607#discussion_r1686027777
+        // TODO: remove this when we can support merging subset of adapters. Ref: https://github.com/ggml-org/llama.cpp/pull/8607#discussion_r1686027777
         static const char * err_no_subset_adapter = "Input adapters do not have the same list of tensors. This is not yet supported. Please merge the adapter one-by-one instead of merging all at once.";
         if (adapters.size() > 1) {
             for (size_t i = 1; i < adapters.size(); ++i) {

tools/perplexity/README.md

@@ -29,7 +29,7 @@ In addition to the KL divergence the following statistics are calculated with `-
 * Mean change in "correct" token probability. Positive values mean the model gets better at prediction, negative values mean it gets worse.
 * Pearson correlation coefficient of the "correct" token probabilites between models.
 * Percentiles of change in "correct" token probability. Positive values mean the model gets better at prediction, negative values mean it gets worse. Can be used to judge noise vs. quality loss from quantization. If the percentiles are symmetric then the quantization is essentially just adding noise. If the negative values are significantly larger than the positive values then this indicates that the model is actually becoming worse from the quantization.
-* The root mean square of the change in token probabilities. If you were to assume that the quantization simply causes Gaussian noise on the token probabilities then this would be the standard deviation of said noise. The uncertainty on the value is calculated that the change in token probabilities follows a Gaussian distribution. Related discussion: https://github.com/ggerganov/llama.cpp/discussions/2875 .
+* The root mean square of the change in token probabilities. If you were to assume that the quantization simply causes Gaussian noise on the token probabilities then this would be the standard deviation of said noise. The uncertainty on the value is calculated that the change in token probabilities follows a Gaussian distribution. Related discussion: https://github.com/ggml-org/llama.cpp/discussions/2875 .
 * Same top p: Percentage of how often the token was assigned the highest probabilites by both models. The uncertainty is calculated from the Gaussian approximation of the binomial distribution.
 
 ## LLaMA 3 8b Scoreboard

tools/server/public_legacy/index-new.html

@@ -1096,7 +1096,7 @@ return html`
           </section>
           <footer>
             <p><${ModelGenerationInfo} /></p>
-            <p>Powered By <a href="https://github.com/ggerganov/llama.cpp#readme" target="_blank">llama.cpp</a> and <a href="https://ggml.ai/" target="_blank">ggml.ai</a></p>
+            <p>Powered By <a href="https://github.com/ggml-org/llama.cpp#readme" target="_blank">llama.cpp</a> and <a href="https://ggml.ai/" target="_blank">ggml.ai</a></p>
           </footer>
         </div>
       `;

tools/server/public_legacy/index.html

@@ -1281,7 +1281,7 @@
 
           <footer>
             <p><${ModelGenerationInfo} /></p>
-            <p>Powered by <a href="https://github.com/ggerganov/llama.cpp">llama.cpp</a> and <a href="https://ggml.ai">ggml.ai</a>.</p>
+            <p>Powered by <a href="https://github.com/ggml-org/llama.cpp">llama.cpp</a> and <a href="https://ggml.ai">ggml.ai</a>.</p>
           </footer>
         </div>
       `;

tools/server/public_legacy/theme-mangotango.css

@@ -1,5 +1,5 @@
 /* Author: Yazan Agha-Schrader */
-/* Inspiration from llama.cpp logo/banner https://github.com/ggerganov/llama.cpp#readme */
+/* Inspiration from llama.cpp logo/banner https://github.com/ggml-org/llama.cpp#readme */
 
 .theme-mangotango {
 

tools/server/themes/buttons-top/index.html

@@ -1032,7 +1032,7 @@
 
           <footer>
             <p><${ModelGenerationInfo} /></p>
-            <p>Powered by <a href="https://github.com/ggerganov/llama.cpp">llama.cpp</a> and <a href="https://ggml.ai">ggml.ai</a>.</p>
+            <p>Powered by <a href="https://github.com/ggml-org/llama.cpp">llama.cpp</a> and <a href="https://ggml.ai">ggml.ai</a>.</p>
           </footer>
         </div>
       `;

tools/server/themes/wild/index.html

@@ -1036,7 +1036,7 @@
 
           <footer>
             <p><${ModelGenerationInfo} /></p>
-            <p>Powered by <a href="https://github.com/ggerganov/llama.cpp">llama.cpp</a> and <a href="https://ggml.ai">ggml.ai</a>.</p>
+            <p>Powered by <a href="https://github.com/ggml-org/llama.cpp">llama.cpp</a> and <a href="https://ggml.ai">ggml.ai</a>.</p>
           </footer>
         </div>
       `;