Update logging documentation

Author: kherud

README.md

@@ -228,6 +228,8 @@ try (LlamaModel model = new LlamaModel(modelParams)) {
 Per default, logs are written to stdout.
 This can be intercepted via the static method `LlamaModel.setLogger(LogFormat, BiConsumer<LogLevel, String>)`. 
 There is text- and JSON-based logging. The default is JSON.
+Note, that text-based logging will include additional output of the GGML backend, while JSON-based logging
+only provides request logs (while still writing GGML messages to stdout).
 To only change the log format while still writing to stdout, `null` can be passed for the callback. 
 Logging can be disabled by passing an empty callback.
 

src/main/java/de/kherud/llama/LlamaModel.java

@@ -99,8 +99,10 @@ public String decode(int[] tokens) {
 
 	/**
 	 * Sets a callback for native llama.cpp log messages.
-	 * Per default, log messages are written to stdout. To only change the log format but keep logging to stdout,
-	 * the given callback can be <code>null</code>.
+	 * Per default, log messages are written in JSON to stdout. Note, that in text mode the callback will be also
+	 * invoked with log messages of the GGML backend, while JSON mode can only access request log messages.
+	 * In JSON mode, GGML messages will still be written to stdout.
+	 * To only change the log format but keep logging to stdout, the given callback can be <code>null</code>.
 	 * To disable logging, pass an empty callback, i.e., <code>(level, msg) -> {}</code>.
 	 *
 	 * @param format the log format to use