diff --git a/benchmark/src/jmh/java/com/knuddels/jtokkit/DataDownloader.java b/benchmark/src/jmh/java/com/knuddels/jtokkit/DataDownloader.java
index 32f75980..f26dcd28 100644
--- a/benchmark/src/jmh/java/com/knuddels/jtokkit/DataDownloader.java
+++ b/benchmark/src/jmh/java/com/knuddels/jtokkit/DataDownloader.java
@@ -178,7 +178,7 @@ public static void main(String[] args) throws Exception {
         }
 
         var totalSize = calculateTotalFileSize(rootFolder);
-        if (totalSize != 99_945_723) {
+        if (totalSize != 99_945_750) {
             throw new AssertionError("Total size did not match expected value, actual: " + totalSize);
         }
     }
diff --git a/lib/src/main/java/com/knuddels/jtokkit/Cl100kParser.java b/lib/src/main/java/com/knuddels/jtokkit/Cl100kParser.java
index c37cc62c..b743fae9 100644
--- a/lib/src/main/java/com/knuddels/jtokkit/Cl100kParser.java
+++ b/lib/src/main/java/com/knuddels/jtokkit/Cl100kParser.java
@@ -41,7 +41,7 @@ static void split(String input, Predicate<ByteArrayList> fragmentConsumer) {
 
             if ((c0 == '\'') && c1 > 0) {
                 if (isShortContraction(c1)) {
-                    // 1) `'[sdtm]` - contractions, such as the suffixes of `he's`, `I'd`, `'tis`, `I'm`
+                    // 1) `'[sdmt]` - contractions, such as the suffixes of `he's`, `I'd`, `'tis`, `I'm`
                     endIndex += 2;
                     finished = fragmentConsumer.test(addUtf8Bytes(input, startIndex, endIndex, utf8Bytes));
                     continue;
diff --git a/lib/src/main/java/com/knuddels/jtokkit/EncodingFactory.java b/lib/src/main/java/com/knuddels/jtokkit/EncodingFactory.java
index 63ac676c..6aa7bd01 100644
--- a/lib/src/main/java/com/knuddels/jtokkit/EncodingFactory.java
+++ b/lib/src/main/java/com/knuddels/jtokkit/EncodingFactory.java
@@ -101,7 +101,7 @@ static Encoding p50kEdit() {
      * @return an {@link Encoding} instance for the cl100k_base encoding
      */
     static Encoding cl100kBase() {
-        // "'(?:[sdmt]|ll|ve|re)|[^\r\n\\p{L}\\p{N}]?+\\p{L}+|\\p{N}{1,3}| ?[^\\s\\p{L}\\p{N}]++[\r\n]*|\\s*[\r\n]|\\s+(?!\\S)|\\s+"
+        // "'(?:[sdmt]|ll|ve|re)|[^\r\n\\p{L}\\p{N}]?+\\p{L}++|\\p{N}{1,3}+| ?[^\\s\\p{L}\\p{N}]++[\r\n]*+|\\s++$|\\s*[\r\n]|\\s+(?!\\S)|\\s"
         Map<byte[], Integer> mergeableRanks = loadMergeableRanks("/com/knuddels/jtokkit/cl100k_base.tiktoken");
         GptBytePairEncodingParams params = new GptBytePairEncodingParams("cl100k_base", null, mergeableRanks, SPECIAL_TOKENS_CL100K_BASE);
         return new Cl100kGptBytePairEncoding(params);
@@ -122,7 +122,7 @@ private static Encoding from50kParameters(
             String fileName,
             Map<String, Integer> specialTokens
     ) {
-        Pattern regex = compileRegex("'(?:[sdmt]|ll|ve|re)| ?\\p{L}+| ?\\p{N}+| ?[^\\s\\p{L}\\p{N}]+|\\s+(?!\\S)|\\s+", false);
+        Pattern regex = compileRegex("'(?:[sdmt]|ll|ve|re)| ?\\p{L}++| ?\\p{N}++| ?[^\\s\\p{L}\\p{N}]++|\\s++$|\\s+(?!\\S)|\\s", false);
         Map<byte[], Integer> mergeableRanks = loadMergeableRanks(fileName);
         GptBytePairEncodingParams params = new GptBytePairEncodingParams(name, regex, mergeableRanks, specialTokens);
         return fromParameters(params);
diff --git a/lib/src/main/java/com/knuddels/jtokkit/TokenEncoder.java b/lib/src/main/java/com/knuddels/jtokkit/TokenEncoder.java
index 82afe1b3..c80b41e5 100644
--- a/lib/src/main/java/com/knuddels/jtokkit/TokenEncoder.java
+++ b/lib/src/main/java/com/knuddels/jtokkit/TokenEncoder.java
@@ -170,11 +170,7 @@ int mergeBytesAndGetTokenCount(ByteArrayWrapper piece, int length, IntArrayList
             ranks.set(nextIndex, DUMMY_RANK);
 
             length--;
-            if (length < 3) {
-                break; // single tokens were already filtered out, let's skip a minimum calculation
-            } else {
-                minRankIndex = getMinRankIndex(ranks);
-            }
+            minRankIndex = getMinRankIndex(ranks);
         }
         assert getMinRankIndex(ranks) < 0;
         return length;
diff --git a/lib/src/main/java/com/knuddels/jtokkit/TokenEncoderLarge.java b/lib/src/main/java/com/knuddels/jtokkit/TokenEncoderLarge.java
index ca3fb411..09c215ac 100644
--- a/lib/src/main/java/com/knuddels/jtokkit/TokenEncoderLarge.java
+++ b/lib/src/main/java/com/knuddels/jtokkit/TokenEncoderLarge.java
@@ -28,7 +28,7 @@ static int calculateTokensLarge(TokenEncoder tokenEncoder, int maxTokenCount, bo
         assert rankMap.containsKey(MAX_RANK);
 
         int tokenCount = match.length();
-        while (tokenCount > 2 && rankMap.size() > 1) {
+        while (rankMap.size() > 1) {
             for (Iterator<RankNode> it = rankMap.pollFirstEntry().getValue().values().iterator(); it.hasNext(); ) {
                 RankNode minNode = it.next();
                 int minRank = minNode.rank;