add grok mixtral

.gitattributes

@@ -39,3 +39,7 @@ vocab/gemma_7b/gemma-7b/tokenizer.model filter=lfs diff=lfs merge=lfs -text
 vocab/gemma_7b/gemma-7b/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 vocab/grok_1/tokenizer.model filter=lfs diff=lfs merge=lfs -text
 vocab/llama3/Meta-Llama-3-70B/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+vocab/mistral_7b/Mistral-7B-v0.1/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+vocab/mistral_7b/Mistral-7B-v0.1/tokenizer.model filter=lfs diff=lfs merge=lfs -text
+vocab/mixtral_8_7b/Mixtral-8x7B-v0.1/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+vocab/mixtral_8_7b/Mixtral-8x7B-v0.1/tokenizer.model filter=lfs diff=lfs merge=lfs -text

README.md

@@ -35,31 +35,26 @@ Check out the configuration reference at https://huggingface.co/docs/hub/spaces-
 https://huggingface.co/spaces/yenniejun/tokenizers-languages
 
 
-## gradio app
-
-- https://arena.lmsys.org/
-
-
-## lang
-
-
-
-## number
 
+## Compress Rate
 
 
-## diff
+在 [cc-100](https://huggingface.co/datasets/cc100) 数据集，每个语言取1万条数据，测试不同tokenizer的压缩率。压缩率指标 `g_bytes/b_tokens`
 
+您可通过以下脚本进行复现 
+```sh
+python utils/compress_rate_util.py 
+```
 
 
 
 
 
-## Compress Rate
+<details> <summary>简体中文压缩率</summary>
+在简体中文数据集 cc100-zh-Hans 计算压缩率 
 
 
-**简介**
-we tokenize in cc-100
+</details>
 
 | tokenizer                   |   vocab_size |   g_bytes/b_tokens |   t_bytes/t_tokens |   b_tokens/g_bytes |
 |:----------------------------|-------------:|-------------------:|-------------------:|-------------------:|

app.py

@@ -59,7 +59,6 @@ with gr.Blocks(css="css/style.css", title="Tokenizer Arena") as demo:
         gr.Markdown("## Input Text")
         dropdown_examples = gr.Dropdown(
             example_types,
-            value="Examples",
             type="index",
             show_label=False,
             container=False,

config.py

@@ -9,4 +9,12 @@ ADD_SPECIAL_TOKEN = False
 LAZY_IMPORT = True
 
 # DEBUG: 设置环境变量 RUST_BACKTRACE=full
-#
+#
+
+default_user_input = """\
+Replace this text in the input field to see how tokenization works.
+Buenos días!
+华为发布Mate60手机。
+ラグビーワールドカップ2023フランス"""
+default_tokenizer_type_1 = "llama3"
+default_tokenizer_type_2 = "gpt4"

evaluation.md

@@ -1,5 +0,0 @@
-
-
-## coverage
-
-rare characters falling back to utf-8 bytes

examples.py

@@ -24,7 +24,7 @@ examples = {
         # ！？｡＂＃＄％＆＇（）＊＋，－／：；＜＝＞＠［＼］＾＿｀｛｜｝～｟｠｢｣､、〃》「」『』【】〔〕〖〗〘〙〚〛〜〝〞〟〰〾〿–—‘’‛“”„‟…‧﹏.
         ["punctuation: ,.:/?+=\"，。！？；【】〔〕〖〗", "gemma_7b", "llama"],  # llama词典有点小
         ["symbol: 🦙❤❥웃유♋☮✊☏☢☚✔☑♚▢♪✈✞÷↑↓▤▥⊙■□▣▽¿─│♥❣▬▫☿Ⓐ ✋✉☣☤", "baichuan", "llama"],
-        ["special: [PAD] [UNK] [CLS] [SEP] [MASK] "],
+        # ["special: [PAD] [UNK] [CLS] [SEP] [MASK] <|endoftext|>", "", ""],
     ],
     "zh": [
         ["空格测试：  2个空格        8个空格", "llama", "chatglm2_6b"],  # chatglm 有blank_n,
@@ -36,15 +36,16 @@ examples = {
 }
 
 more_examples = [
-    # bert VS clue
     # bert系列
-    ("bert_base_cased", "bert_base_uncased", ""),  # # clue VS kplug， bert VS clue
-    ("bert_base_cased", "clue", ""),
+    ("bert_base_cased", "bert_base_uncased", "", ""),  # # clue VS kplug， bert VS clue
+    ("bert_base_cased", "clue", "", "增加了[]()"),
+    ("clue", "kplug", "", ""),
 
     # llama系列 (基于sentencepiece)
     ("baichuan", "baichuan2", "baichuan2支持多空格   ，多个换行\n\n\n，do not add dummy prefix as Baichuan1"),
     ("llama", "baichuan2", "baichuan2支持多空格   ，多个换行\n\n"),
     ("llama", "chinese_llama2", ""),
+    ("llama", "llama3", "扩充词典"),
     ("chinese_llama", "chinese_llama2", ""),
 
     # glm系列 （基于sentencepiece）
@@ -72,7 +73,7 @@ def example_fn(example_idx):
 def get_more_example():
     import urllib.parse
     url_prefix = "https://huggingface.co/spaces/eson/tokenizer-arena"
-    for tokenizer1, tokenizer2, text in more_examples:
+    for tokenizer1, tokenizer2, text, comment in more_examples:
         full_url = f'{url_prefix}?tokenizer1={tokenizer1}&tokenizer2={tokenizer2}&text={urllib.parse.quote(text)}'
         print(full_url)
 

js/onload.js

@@ -3,10 +3,16 @@ function() {
     //$("footer a")["href"] = "https://github.com/xu-song/tokenizer-arena/issues"
     //$("footer a").childNodes[0].textContent ="Send Feedback"
 
-    document.querySelectorAll("footer a")[0].childNodes[0].textContent ="Send Feedback";  //  🤔Reporting Issues
+
+//    <a href="https://gradio.app" class="built-with svelte-16bt5n8" target="_blank" rel="noreferrer">
+//          Built with Gradio <img src="http://127.0.0.1:7860/assets/logo-3707f936.svg" alt="logo" class="svelte-16bt5n8">
+//    </a>
+    console.log("nice")
+    document.querySelectorAll("footer a")[0].childNodes[0].textContent ="Send Feedback";  //  🤔Reporting Issues, 💬Leave comments
     document.querySelectorAll("footer a")[0].href = "https://github.com/xu-song/tokenizer-arena/issues";
 
     // download button
 
     // API
+    return 'Animation created';
 }

util.py

@@ -117,14 +117,6 @@ def get_overlap_token_size(tokenizer_type_1, tokenizer_type_2):
     return overlap_token_size, overlap_token_size
 
 
-default_user_input = """Replace this text in the input field to see how tokenization works.
-Buenos días!
-华为发布Mate60手机。
-ラグビーワールドカップ2023フランス"""
-default_tokenizer_type_1 = "llama"
-# default_tokenizer_type_2 = "internlm_chat_7b"
-default_tokenizer_type_2 = "gpt_35_turbo"
-
 
 def on_load(url_params, request: gr.Request):
     """
@@ -148,9 +140,9 @@ def on_load(url_params, request: gr.Request):
         # if "referer" in request.headers:   # not work for huggingface-space
         #     url_params = parse_qs(urlparse(request.headers["referer"]).query)
         #     url_params = {k: v[0] for k, v in url_params.items() if len(v) > 0}
-        tokenizer_type_1 = url_params.get("tokenizer1", default_tokenizer_type_1)
-        tokenizer_type_2 = url_params.get("tokenizer2", default_tokenizer_type_2)
-        text = url_params.get("text", default_user_input)
+        tokenizer_type_1 = url_params.get("tokenizer1", config.default_tokenizer_type_1)
+        tokenizer_type_2 = url_params.get("tokenizer2", config.default_tokenizer_type_2)
+        text = url_params.get("text", config.default_user_input)
         logger.info(f"client_ip: {client_ip}; params: {url_params}")
     return text, tokenizer_type_1, tokenizer_type_2
 

utils/compress_rate_util.py

@@ -99,7 +99,7 @@ def pprint(stats):
         table.append(columns)
     df = pd.DataFrame(table)
     # print(df.to_markdown(index=False, tablefmt='fancy_grid'))
-    logger.info(df.to_markdown(index=False))
+    logger.info("\n{df.to_markdown(index=False)}")
     return
 
 
@@ -167,8 +167,8 @@ def main():
     for lang in ["en", "zh-Hans"]:
         print("###" * 10 + lang)
 
-        for tokenizer_name in ['llama', 'llama2', 'llama3']:
-            # for tokenizer_name in all_tokenizers:
+        # for tokenizer_name in ['llama', 'llama2', 'llama3']:
+        for tokenizer_name in all_tokenizers:
             tokenizer = load_tokener(tokenizer_name)
             stat = tokenize_corpus(tokenizer, lang)
             # ["qwen1_5_14b_chat", "gpt_35_turbo",]:

vocab/code_davinci_002/__init__.py

@@ -1,3 +1,5 @@
 
+import copy
+from vocab.text_davinci_003 import tokenizer as base_tokenizer
 
-from vocab.text_davinci_003 import tokenizer
+tokenizer = copy.copy(base_tokenizer)

vocab/dbrx_instruct/__init__.py

@@ -0,0 +1,8 @@
+
+import os
+from transformers import AutoTokenizer
+
+CURRENT_DIR = os.path.dirname(os.path.abspath(__file__))
+TOKENIZER_DIR = os.path.join(CURRENT_DIR, "dbrx-instruct")
+tokenizer = AutoTokenizer.from_pretrained(TOKENIZER_DIR, trust_remote_code=True)
+

vocab/dbrx_instruct/dbrx-instruct/added_tokens.json

@@ -0,0 +1,5 @@
+{
+  "<|im_end|>": 100279,
+  "<|im_start|>": 100278,
+  "<|pad|>": 100277
+}

vocab/dbrx_instruct/dbrx-instruct/special_tokens_map.json

@@ -0,0 +1,34 @@
+{
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>"
+  ],
+  "bos_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<|pad|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

vocab/dbrx_instruct/dbrx-instruct/tiktoken.py

@@ -0,0 +1,374 @@
+"""Dbrx tokenizer."""
+
+from functools import lru_cache
+from typing import Any, Dict, List, Optional, Tuple
+
+from transformers import PreTrainedTokenizer
+
+
+def dbrx_system_prompt():
+    # This is inspired by the Claude3 prompt.
+    # source: https://twitter.com/AmandaAskell/status/1765207842993434880
+    # Identity and knowledge
+    prompt = 'You are DBRX, created by Databricks. You were last updated in December 2023. You answer questions based on information available up to that point.\n'
+    prompt += 'YOU PROVIDE SHORT RESPONSES TO SHORT QUESTIONS OR STATEMENTS, but provide thorough responses to more complex and open-ended questions.\n'
+    # Capabilities (and reminder to use ``` for JSON blocks and tables, which it can forget). Also a reminder that it can't browse the internet or run code.
+    prompt += 'You assist with various tasks, from writing to coding (using markdown for code blocks — remember to use ``` with code, JSON, and tables).\n'
+    prompt += '(You do not have real-time data access or code execution capabilities. '
+    # Ethical guidelines
+    prompt += 'You avoid stereotyping and provide balanced perspectives on controversial topics. '
+    # Data: the model doesn't know what it was trained on; it thinks that everything that it is aware of was in its training data. This is a reminder that it wasn't.
+    # We also encourage it not to try to generate lyrics or poems
+    prompt += 'You do not provide song lyrics, poems, or news articles and do not divulge details of your training data.)\n'
+    # The model really wants to talk about its system prompt, to the point where it is annoying, so encourage it not to
+    prompt += 'This is your system prompt, guiding your responses. Do not reference it, just respond to the user. If you find yourself talking about this message, stop. You should be responding appropriately and usually that means not mentioning this.\n'
+    prompt += 'You do not mention any of this information about yourself unless the information is directly pertinent to the user\\\'s query.'.upper()
+    return prompt
+
+
+# Taken from
+# https://github.com/huggingface/transformers/blob/8aca43bdb3cb9a5020f6d57589d85679dc873b1c/src/transformers/models/gpt2/tokenization_gpt2.py#L62-L84
+@lru_cache()
+def bytes_to_unicode():
+    """Returns list of utf-8 byte and a mapping to unicode strings.
+
+    We specifically avoids mapping to whitespace/control characters the bpe code
+    barfs on.
+
+    The reversible bpe codes work on unicode strings. This means you need a
+    large # of unicode characters in your vocab if you want to avoid UNKs. When
+    you're at something like a 10B token dataset you end up needing around 5K
+    for decent coverage. This is a significant percentage of your normal, say,
+    32K bpe vocab. To avoid that, we want lookup tables between utf-8 bytes and
+    unicode strings.
+    """
+    bs = (list(range(ord('!'),
+                     ord('~') + 1)) + list(range(ord('¡'),
+                                                 ord('¬') + 1)) +
+          list(range(ord('®'),
+                     ord('ÿ') + 1)))
+    cs = bs[:]
+    n = 0
+    for b in range(2**8):
+        if b not in bs:
+            bs.append(b)
+            cs.append(2**8 + n)
+            n += 1
+    cs = [chr(n) for n in cs]
+    return dict(zip(bs, cs))
+
+
+class TiktokenTokenizerWrapper(PreTrainedTokenizer):
+    """A thin wrapper around tiktoken to make it compatible with Hugging Face.
+
+    tokenizers.
+
+    See HuggingFace for further documentation on general tokenizer methods.
+    """
+
+    model_input_names = ['input_ids', 'attention_mask']
+
+    def __init__(self,
+                 model_name: Optional[str] = None,
+                 encoding_name: Optional[str] = None,
+                 add_bos_token: bool = False,
+                 add_eos_token: bool = False,
+                 use_default_system_prompt: bool = False,
+                 unk_token: Optional[str] = '<|endoftext|>',
+                 eos_token: Optional[str] = '<|endoftext|>',
+                 bos_token: Optional[str] = '<|endoftext|>',
+                 pad_token: Optional[str] = None,
+                 errors: str = 'replace',
+                 **kwargs: Any):
+        """Constructor creates a tiktoken tokenizer to use as the underlying.
+
+        tokenizer.
+
+        Args:
+            model_name (Optional[str], optional): The name of the model to load from tiktoken. Defaults to None.
+                Either model_name or encoding_name must be set, but not both.
+            encoding_name (Optional[str], optional): The name of the encoding to load from tiktoken. Defaults to None.
+                Either model_name or encoding_name must be set, but not both.
+            add_bos_token (bool, optional): Whether to add bos tokens. Defaults to False.
+            add_eos_token (bool, optional): Whether to add eos tokens. Defaults to False.
+            use_default_system_prompt (bool, optional): Use the default system prompt or not. Defaults to False.
+            unk_token (Optional[str], optional): The unk token. Defaults to '<|endoftext|>'.
+            eos_token (Optional[str], optional): The eos token. Defaults to '<|endoftext|>'.
+            bos_token (Optional[str], optional): The bos token. Defaults to '<|endoftext|>'.
+            pad_token (Optional[str], optional): The pad token. Defaults to None.
+            errors (str, optional): Paradigm to follow when decoding bytes to UTF-8. See
+                [bytes.decode](https://docs.python.org/3/library/stdtypes.html#bytes.decode) for more information.
+                Defaults to `"replace"`.
+        """
+        try:
+            import tiktoken
+        except:
+            raise ImportError(
+                'You need to install tiktoken to use TiktokenTokenizerWrapper.')
+
+        # Workaround to make tiktokenizer picklable.
+        # https://github.com/huggingface/datasets/issues/5536#issuecomment-1682309347
+        # There is an open PR from HF to add this to tiktoken: https://github.com/openai/tiktoken/pull/181
+        import copyreg
+        import functools
+
+        from tiktoken import Encoding  # type: ignore (thirdParty)
+
+        def pickle_Encoding(enc: Encoding):
+            return (functools.partial(Encoding,
+                                      enc.name,
+                                      pat_str=enc._pat_str,
+                                      mergeable_ranks=enc._mergeable_ranks,
+                                      special_tokens=enc._special_tokens), ())
+
+        copyreg.pickle(Encoding, pickle_Encoding)
+
+        if model_name is not None and encoding_name is not None:
+            raise ValueError(
+                'You need to specify either model_name or encoding_name, not both.'
+            )
+
+        self.model_name = model_name
+        self.encoding_name = encoding_name
+
+        if self.model_name is not None:
+            self.encoding = tiktoken.encoding_for_model(  # type: ignore (thirdParty)
+                self.model_name)
+        elif self.encoding_name is not None:
+            self.encoding = tiktoken.get_encoding(  # type: ignore (thirdParty)
+                self.encoding_name)
+        else:
+            raise ValueError(
+                'You need to specify either model_name or encoding_name.')
+
+        self.add_bos_token = add_bos_token
+        self.add_eos_token = add_eos_token
+        self.use_default_system_prompt = use_default_system_prompt
+
+        self.byte_encoder = bytes_to_unicode()
+        self.byte_decoder = {v: k for k, v in self.byte_encoder.items()}
+        self.errors = errors
+
+        self.decoder: Dict[int, str] = {}
+        for i in range(self.encoding.n_vocab):
+            try:
+                self.encoding.decode_single_token_bytes(i)
+            except KeyError:
+                continue
+            # Taken from
+            # https://gist.github.com/xenova/a452a6474428de0182b17605a98631ee
+            decoding = ''.join([
+                bytes_to_unicode()[ord(char)] for char in
+                self.encoding.decode_single_token_bytes(i).decode('latin-1')
+            ])
+            self.decoder[i] = decoding
+
+        self.encoder: Dict[str, int] = {}
+        for i in range(self.encoding.n_vocab):
+            if i in self.decoder:
+                self.encoder[self.decoder[i]] = i
+
+        super().__init__(model_name=model_name,
+                         encoding_name=encoding_name,
+                         add_bos_token=add_bos_token,
+                         add_eos_token=add_eos_token,
+                         use_default_system_prompt=use_default_system_prompt,
+                         unk_token=unk_token,
+                         eos_token=eos_token,
+                         bos_token=bos_token,
+                         pad_token=pad_token,
+                         errors=errors,
+                         **kwargs)
+
+    @property
+    def vocab_size(self) -> int:
+        """Returns vocab size."""
+        return self.encoding.n_vocab
+
+    @property
+    def is_fast(self) -> bool:
+        return False
+
+    @property
+    def default_chat_template(self):
+        """Chat ML Template for User/Assistant.
+
+        Pinning default Chat ML template in case defaults change.
+        """
+        template = (
+            "{% if messages[0]['role'] == 'system' %}"
+            '{% set loop_messages = messages[1:] %}'
+            "{% set system_message = messages[0]['content'] %}"
+            "{% elif USE_DEFAULT_PROMPT == true and not 'system' in messages[0]['role'] %}"
+            '{% set loop_messages = messages %}'
+            "{% set system_message = 'DEFAULT_SYSTEM_PROMPT' %}"
+            '{% else %}'
+            '{% set loop_messages = messages %}'
+            '{% set system_message = false %}'
+            '{% endif %}'
+            '{% for message in loop_messages %}'
+            '{% if loop.index0 == 0 %}'
+            '{% if system_message != false %}'
+            "{{ '<|im_start|>system\n' + system_message.strip() + '<|im_end|>\n'}}"
+            '{% endif %}'
+            "{{ '<|im_start|>' + message['role'] + '\n' + message['content'] + '<|im_end|>' }}"
+            '{% else %}'
+            "{{ '\n' + '<|im_start|>' + message['role'] + '\n' + message['content'] + '<|im_end|>' }}"
+            '{% endif %}'
+            '{% if (add_generation_prompt == true and loop.last) %}'
+            "{{ '\n' + '<|im_start|>' + 'assistant' + '\n' }}"
+            '{% endif %}'
+            '{% endfor %}')
+        template = template.replace(
+            'USE_DEFAULT_PROMPT',
+            'true' if self.use_default_system_prompt else 'false')
+        template = template.replace('DEFAULT_SYSTEM_PROMPT',
+                                    dbrx_system_prompt())
+        return template
+
+    def get_vocab(self) -> Dict[str, int]:
+        """Returns vocab as a dict."""
+        # As far as I can tell, we don't require get_vocab to completely work,
+        # but when using additional_special_tokens, Hugging Face determines the next
+        # token index to add with len(self.get_vocab()) so we need the _size_ of this dictionary to be correct.
+        vocab_clone = self.encoder.copy()
+        extra_id_index = 0
+        candidate_extra_id = f'<extra_id_{extra_id_index}>'
+        indices_to_fill_in = {i for i in range(self.vocab_size)} - set(
+            vocab_clone.values())
+
+        # Add enough indices to make get_vocab() the right length
+        for index_to_add in indices_to_fill_in:
+            # Make sure we don't overwrite a token that already exists
+            while candidate_extra_id in vocab_clone:
+                extra_id_index += 1
+                candidate_extra_id = f'<extra_id_{extra_id_index}>'
+
+            # Get an index to add and add the item
+            vocab_clone[candidate_extra_id] = index_to_add
+
+        return dict(vocab_clone, **self.added_tokens_encoder)
+
+    def _tokenize(self, text: str) -> List[str]:
+        """Returns a tokenized string."""
+        if not isinstance(text, str):
+            raise ValueError(
+                f'Expected a string input to _tokenize but got {type(text)}.')
+
+        tokens = [
+            self.decoder[t]
+            for t in self.encoding.encode(text, allowed_special='all')
+        ]
+
+        return tokens
+
+    def _convert_token_to_id(self, token: str) -> Optional[int]:
+        """Converts a token (str) in an id using the vocab."""
+        return self.encoder.get(token, self.encoder.get(self.unk_token))
+
+    def _convert_id_to_token(self, index: int) -> Optional[str]:
+        """Converts an index (integer) in a token (str) using the vocab."""
+        # For tokens in either the gap in ids in the tokenizer, or beyond the range of the tokenizer,
+        # we return empty string. This matches the behavior of Hugging Face fast tokenizers,
+        # but not slow tokenizers.
+        return self.decoder.get(index, '')
+
+    def convert_tokens_to_string(self, tokens: List[str]) -> str:
+        """Converts a sequence of tokens (string) in a single string."""
+        text = ''.join(tokens)
+        text = bytearray([self.byte_decoder[c] for c in text
+                         ]).decode('utf-8', errors=self.errors)
+        return text
+
+    def build_inputs_with_special_tokens(
+            self,
+            token_ids_0: List[int],
+            token_ids_1: Optional[List[int]] = None) -> List[int]:
+        bos_token_id = [self.bos_token_id] if self.add_bos_token else []
+        eos_token_id = [self.eos_token_id] if self.add_eos_token else []
+
+        output = bos_token_id + token_ids_0 + eos_token_id
+
+        if token_ids_1 is not None:
+            output = output + bos_token_id + token_ids_1 + eos_token_id
+
+        return output
+
+    def get_special_tokens_mask(
+            self,
+            token_ids_0: List[int],
+            token_ids_1: Optional[List[int]] = None,
+            already_has_special_tokens: bool = False) -> List[int]:
+        """Retrieves sequence ids from a token list that has no special tokens.
+
+        Function copied from
+        https://github.com/huggingface/transformers/blob/e3a4bd2bee212a2d0fd9f03b27fe7bfc1debe42d/src/transformers/models/gpt2/tokenization_gpt2.py#L265-L295
+
+        added. This method is called when adding special tokens using the
+        tokenizer `prepare_for_model` or `encode_plus` methods.
+
+        Args:
+            token_ids_0 (`List[int]`):
+                List of IDs.
+            token_ids_1 (`List[int]`, *optional*):
+                Optional second list of IDs for sequence pairs.
+            already_has_special_tokens (`bool`, *optional*, defaults to `False`):
+                Whether or not the token list is already formatted with special tokens for the model.
+
+        Returns:
+            `List[int]`: A list of integers in the range [0, 1]: 1 for a special token, 0 for a sequence token.
+        """
+        if already_has_special_tokens:
+            return super().get_special_tokens_mask(
+                token_ids_0=token_ids_0,
+                token_ids_1=token_ids_1,
+                already_has_special_tokens=True)
+
+        bos_token_id = [1] if self.add_bos_token else []
+        eos_token_id = [1] if self.add_eos_token else []
+
+        if token_ids_1 is None:
+            return bos_token_id + ([0] * len(token_ids_0)) + eos_token_id
+        return (bos_token_id + ([0] * len(token_ids_0)) + eos_token_id +
+                bos_token_id + ([0] * len(token_ids_1)) + eos_token_id)
+
+    def create_token_type_ids_from_sequences(
+            self,
+            token_ids_0: List[int],
+            token_ids_1: Optional[List[int]] = None) -> List[int]:
+        sep = [self.sep_token_id]
+
+        if token_ids_1 is None:
+            return len(token_ids_0 + sep) * [0]
+        return len(token_ids_0 + sep) * [0] + len(token_ids_1 + sep) * [1]
+
+    def save_vocabulary(self,
+                        save_directory: str,
+                        filename_prefix: Optional[str] = None) -> Tuple[str]:
+
+        # ignore the below type to keep the original signature
+        # we are knowingly breaking the signature here, although not 100% certain
+        # it doesn't have side effects
+        # There is some code in huggingface that calls this function to get the vocab files,
+        # but it doesn't seem to access them (or at least checks for their existence
+        # before accessing them)
+        return (None, None)  # type: ignore
+
+    def sanitize_special_tokens(self) -> int:
+        """Make sure that all the special tokens attributes of the tokenizer.
+
+        (`tokenizer.mask_token`, `tokenizer.cls_token`, etc.) are in the
+        vocabulary.
+
+        Add the missing ones to the vocabulary if needed.
+
+        Return:
+            `int`: The number of tokens added in the vocabulary during the operation.
+        """
+        actual_new_tokens = []
+        for token in self.all_special_tokens_extended:
+            encoded = self.encoding.encode(token, allowed_special='all')
+            if len(encoded) > 1:
+                actual_new_tokens.append(token)
+
+        return self.add_tokens(actual_new_tokens, special_tokens=True)

vocab/dbrx_instruct/dbrx-instruct/tokenizer_config.json

@@ -0,0 +1,60 @@
+{
+  "add_bos_token": false,
+  "add_eos_token": false,
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "100257": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "100277": {
+      "content": "<|pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "100278": {
+      "content": "<|im_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "100279": {
+      "content": "<|im_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>"
+  ],
+  "auto_map": {
+    "AutoTokenizer": [
+      "tiktoken.TiktokenTokenizerWrapper",
+      null
+    ]
+  },
+  "bos_token": "<|endoftext|>",
+  "clean_up_tokenization_spaces": true,
+  "encoding_name": null,
+  "eos_token": "<|endoftext|>",
+  "errors": "replace",
+  "model_max_length": 1000000000000000019884624838656,
+  "model_name": "gpt-4",
+  "pad_token": "<|pad|>",
+  "tokenizer_class": "TiktokenTokenizerWrapper",
+  "unk_token": "<|endoftext|>",
+  "use_default_system_prompt": true
+}

vocab/grok_1/__init__.py

@@ -0,0 +1,21 @@
+"""
+
+## reference
+https://github.com/xai-org/grok-1/blob/main/run.py
+
+        vocab_size=128 * 1024,
+        pad_token=0,
+        eos_token=2,
+"""
+
+
+import os
+import sentencepiece
+from tokenizer import sptokenizer_patch
+
+CURRENT_DIR = os.path.dirname(os.path.abspath(__file__))
+MODEL_FILE = os.path.join(CURRENT_DIR, "tokenizer.model")
+
+tokenizer = sentencepiece.SentencePieceProcessor(model_file=MODEL_FILE)
+
+# print(tokenizer.decode([1,2,3], skip_special_tokens=True))

vocab/grok_1/tokenizer.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5c69d7cbad192fa2c9d14e2a77fbfdc11597c68907b043ddb0260e3d28eddd7f
+size 2229219

vocab/mistral_7b/__init__.py

@@ -1,5 +1,13 @@
 
+import os
+import config
+from transformers import AutoTokenizer
 
 
-from transformers import AutoTokenizer
-tokenizer = AutoTokenizer.from_pretrained("mistralai/Mistral-7B-v0.1", trust_remote_code=True)
+
+if config.USE_REMOTE:
+    tokenizer = AutoTokenizer.from_pretrained("mistralai/Mistral-7B-v0.1", trust_remote_code=True)
+else:
+    CURRENT_DIR = os.path.dirname(os.path.abspath(__file__))
+    TOKENIZER_DIR = os.path.join(CURRENT_DIR, "Mistral-7B-v0.1")
+    tokenizer = AutoTokenizer.from_pretrained(TOKENIZER_DIR, trust_remote_code=True)

vocab/mixtral_8_7b/Mixtral-8x7B-v0.1/special_tokens_map.json

@@ -0,0 +1,5 @@
+{
+  "bos_token": "<s>",
+  "eos_token": "</s>",
+  "unk_token": "<unk>"
+}

vocab/mixtral_8_7b/Mixtral-8x7B-v0.1/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fc4f0bd70b3709312d9d1d9e5ba674794b6bc5abc17429897a540f93882f25fc
+size 1795303

vocab/mixtral_8_7b/Mixtral-8x7B-v0.1/tokenizer.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dadfd56d766715c61d2ef780a525ab43b8e6da4de6865bda3d95fdef5e134055
+size 493443

vocab/mixtral_8_7b/Mixtral-8x7B-v0.1/tokenizer_config.json

@@ -0,0 +1,42 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "additional_special_tokens": [],
+  "bos_token": "<s>",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "</s>",
+  "legacy": true,
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": null,
+  "sp_model_kwargs": {},
+  "spaces_between_special_tokens": false,
+  "tokenizer_class": "LlamaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}

vocab/mixtral_8_7b/__init__.py

@@ -1,2 +1,14 @@
+
+
+import os
+import config
 from transformers import AutoTokenizer
-tokenizer = AutoTokenizer.from_pretrained("mistralai/Mixtral-8x7B-v0.1", trust_remote_code=True)
+
+
+if config.USE_REMOTE:
+    tokenizer = AutoTokenizer.from_pretrained("mistralai/Mixtral-8x7B-v0.1", trust_remote_code=True)
+else:
+    CURRENT_DIR = os.path.dirname(os.path.abspath(__file__))
+    TOKENIZER_DIR = os.path.join(CURRENT_DIR, "Mixtral-8x7B-v0.1")
+    tokenizer = AutoTokenizer.from_pretrained(TOKENIZER_DIR, trust_remote_code=True)
+

vocab/text_davinci_003/__init__.py

@@ -18,7 +18,6 @@ import tiktoken
 import tokenizer.tiktoken_patch
 
 tokenizer = tiktoken.encoding_for_model('text-davinci-003')
-tokenizer.vocab_size = tokenizer.n_vocab
 
 tokenizer.comments = ""
 tokenizer.reversible = True