meta-llama

/

Meta-Llama-3-8B

like 6.14k

Follow

Meta Llama 37.4k

Wait....Llama3 switched to TikToken tokenizer? Am I correct?

Ok, yeah, confirming that Llama3 has switched to tiktoken from sentencepiece. Here is a wrapper code you may use (still WIP but seems to work):

import os
import sentencepiece as spm
import tiktoken
from tiktoken.load import load_tiktoken_bpe
from pathlib import Path
from typing import Dict

class TokenizerInterface:
    def __init__(self, model_path):
        self.model_path = model_path

    def encode(self, text):
        raise NotImplementedError("This method should be overridden by subclasses.")

    def decode(self, tokens):
        raise NotImplementedError("This method should be overridden by subclasses.")

    def bos_id(self):
        raise NotImplementedError("This method should be overridden by subclasses.")

    def eos_id(self):
        raise NotImplementedError("This method should be overridden by subclasses.")

class SentencePieceWrapper(TokenizerInterface):
    def __init__(self, model_path):
        super().__init__(model_path)
        self.processor = spm.SentencePieceProcessor(str(model_path))

    def encode(self, text):
        return self.processor.EncodeAsIds(text)

    def decode(self, tokens):
        return self.processor.DecodeIds(tokens)

    def bos_id(self):
        return self.processor.bos_id()

    def eos_id(self):
        return self.processor.eos_id()

class TiktokenWrapper(TokenizerInterface):
    """
    Tokenizing and encoding/decoding text using the Tiktoken tokenizer.
    """

    special_tokens: Dict[str, int]

    num_reserved_special_tokens = 256

    pat_str = r"(?i:'s|'t|'re|'ve|'m|'ll|'d)|[^\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+"  # noqa: E501

    def __init__(self, model_path):
        super().__init__(model_path)
        assert os.path.isfile(model_path), str(model_path)
        mergeable_ranks = load_tiktoken_bpe(str(model_path))
        num_base_tokens = len(mergeable_ranks)
        special_tokens = [
            "<|begin_of_text|>",
            "<|end_of_text|>",
            "<|reserved_special_token_0|>",
            "<|reserved_special_token_1|>",
            "<|reserved_special_token_2|>",
            "<|reserved_special_token_3|>",
            "<|start_header_id|>",
            "<|end_header_id|>",
            "<|reserved_special_token_4|>",
            "<|eot_id|>",  # end of turn
        ] + [
            f"<|reserved_special_token_{i}|>"
            for i in range(5, self.num_reserved_special_tokens - 5)
        ]
        self.special_tokens = {
            token: num_base_tokens + i for i, token in enumerate(special_tokens)
        }
        self.model = tiktoken.Encoding(
            name=Path(model_path).name,
            pat_str=self.pat_str,
            mergeable_ranks=mergeable_ranks,
            special_tokens=self.special_tokens,
        )
        # BOS / EOS token IDs
        self._bos_id: int = self.special_tokens["<|begin_of_text|>"]
        self._eos_id: int = self.special_tokens["<|end_of_text|>"]

    def encode(self, text):
        return self.model.encode(text)

    def decode(self, tokens):
        return self.model.decode(tokens)

    def bos_id(self):
        return self._bos_id

    def eos_id(self):
        return self._eos_id

def get_tokenizer(tokenizer_model_path, model_name):
    """
    Factory function to get the appropriate tokenizer based on the model name.
    
    Args:
    - tokenizer_model_path (str): The file path to the tokenizer model.
    - model_name (str): The name of the model, used to determine the tokenizer type.

    Returns:
    - TokenizerInterface: An instance of a tokenizer.
    """
    if "Llama-3" in str(model_name):
        return TiktokenWrapper(tokenizer_model_path)
    else:
        return SentencePieceWrapper(tokenizer_model_path)

Hmmm, so run the demos it need a file named tokenizer.model. Someone says the file is no longer there.
At the end of the day this closed discussion didn't result if producing a tokenizer.model file.
What am I missing.

There is a tokenizer.model file in 'original' sub-dir. The thing is - you need to use TikTokenizer to load it instead of the previous SentencePiece one. The code above shows you how to handle it.

Hello.

I wonder if there is an article / document that describes the tiktoken algorithm in plain English.