llama : fix BPE pre-tokenization (#6920)

* merged the changes from deepseeker models to main branch

* Moved regex patterns to unicode.cpp and updated unicode.h

* Moved header files

* Resolved issues

* added and refactored unicode_regex_split and related functions

* Updated/merged the deepseek coder pr

* Refactored code

* Adding unicode regex mappings

* Adding unicode regex function

* Added needed functionality, testing remains

* Fixed issues

* Fixed issue with gpt2 regex custom preprocessor

* unicode : fix? unicode_wstring_to_utf8

* lint : fix whitespaces

* tests : add tokenizer tests for numbers

* unicode : remove redundant headers

* tests : remove and rename tokenizer test scripts

* tests : add sample usage

* gguf-py : reader prints warnings on duplicate keys

* llama : towards llama3 tokenization support (wip)

* unicode : shot in the dark to fix tests on Windows

* unicode : first try custom implementations

* convert : add "tokenizer.ggml.pre" GGUF KV (wip)

* llama : use new pre-tokenizer type

* convert : fix pre-tokenizer type writing

* lint : fix

* make : add test-tokenizer-0-llama-v3

* wip

* models : add llama v3 vocab file

* llama : adapt punctuation regex + add llama 3 regex

* minor

* unicode : set bomb

* unicode : set bomb

* unicode : always use std::wregex

* unicode : support \p{N}, \p{L} and \p{P} natively

* unicode : try fix windows

* unicode : category support via std::regex

* unicode : clean-up

* unicode : simplify

* convert : add convert-hf-to-gguf-update.py

ggml-ci

* lint : update

* convert : add falcon

ggml-ci

* unicode : normalize signatures

* lint : fix

* lint : fix

* convert : remove unused functions

* convert : add comments

* convert : exercise contractions

ggml-ci

* lint : fix

* cmake : refactor test targets

* tests : refactor vocab tests

ggml-ci

* tests : add more vocabs and tests

ggml-ci

* unicode : cleanup

* scripts : ignore new update script in check-requirements.sh

* models : add phi-3, mpt, gpt-2, starcoder

* tests : disable obsolete

ggml-ci

* tests : use faster bpe test

ggml-ci

* llama : more prominent warning for old BPE models

* tests : disable test-tokenizer-1-bpe due to slowness

ggml-ci

---------

Co-authored-by: Jaggzh <jaggz.h@gmail.com>
Co-authored-by: Kazim Abrar Mahi <kazimabrarmahi135@gmail.com>

gguf-py/gguf/constants.py

@@ -72,6 +72,7 @@ class Keys:
 
     class Tokenizer:
         MODEL            = "tokenizer.ggml.model"
+        PRE              = "tokenizer.ggml.pre"
         LIST             = "tokenizer.ggml.tokens"
         TOKEN_TYPE       = "tokenizer.ggml.token_type"
         TOKEN_TYPE_COUNT = "tokenizer.ggml.token_type_count"  # for BERT-style token types
@@ -940,6 +941,7 @@ KEY_SSM_TIME_STEP_RANK = Keys.SSM.TIME_STEP_RANK
 
 # tokenization
 KEY_TOKENIZER_MODEL      = Keys.Tokenizer.MODEL
+KEY_TOKENIZER_PRE        = Keys.Tokenizer.PRE
 KEY_TOKENIZER_LIST       = Keys.Tokenizer.LIST
 KEY_TOKENIZER_TOKEN_TYPE = Keys.Tokenizer.TOKEN_TYPE
 KEY_TOKENIZER_SCORES     = Keys.Tokenizer.SCORES

gguf-py/gguf/gguf_reader.py

@@ -139,8 +139,13 @@ class GGUFReader:
 
     def _push_field(self, field: ReaderField, skip_sum: bool = False) -> int:
         if field.name in self.fields:
-            raise KeyError(f'Duplicate {field.name} already in list at offset {field.offset}')
-        self.fields[field.name] = field
+            # TODO: add option to generate error on duplicate keys
+            # raise KeyError(f'Duplicate {field.name} already in list at offset {field.offset}')
+
+            print(f'Warning: Duplicate key {field.name} at offset {field.offset}')
+            self.fields[field.name + '_{}'.format(field.offset)] = field
+        else:
+            self.fields[field.name] = field
         return 0 if skip_sum else sum(int(part.nbytes) for part in field.parts)
 
     def _get_str(self, offset: int) -> tuple[npt.NDArray[np.uint64], npt.NDArray[np.uint8]]:

gguf-py/gguf/gguf_writer.py

@@ -427,6 +427,9 @@ class GGUFWriter:
     def add_tokenizer_model(self, model: str) -> None:
         self.add_string(Keys.Tokenizer.MODEL, model)
 
+    def add_tokenizer_pre(self, pre: str) -> None:
+        self.add_string(Keys.Tokenizer.PRE, pre)
+
     def add_token_list(self, tokens: Sequence[str] | Sequence[bytes] | Sequence[bytearray]) -> None:
         self.add_array(Keys.Tokenizer.LIST, tokens)