better handling for tokenizers like flan that don't have a bos token

.github/workflows/base.yml

@@ -26,7 +26,7 @@ jobs:
             pytorch: 2.0.0
             axolotl_extras:
           - cuda: "117"
-            cuda_version: 11.7.1
+            cuda_version: 11.7.0
             python_version: "3.9"
             pytorch: 1.13.1
             axolotl_extras:

.github/workflows/main.yml

@@ -30,7 +30,7 @@ jobs:
             pytorch: 2.0.0
             axolotl_extras: gptq
           - cuda: cu117
-            cuda_version: 11.7.1
+            cuda_version: 11.7.0
             python_version: "3.9"
             pytorch: 1.13.1
             axolotl_extras:
@@ -85,7 +85,7 @@ jobs:
             pytorch: 2.0.0
             axolotl_extras: gptq
           - cuda: cu117
-            cuda_version: 11.7.1
+            cuda_version: 11.7.0
             python_version: "3.9"
             pytorch: 1.13.1
             axolotl_extras:

.pre-commit-config.yaml

@@ -1,5 +1,5 @@
 default_language_version:
-    python: python3
+    python: python3.9
 
 repos:
 -   repo: https://github.com/pre-commit/pre-commit-hooks

README.md

@@ -302,8 +302,6 @@ model_type: AutoModelForCausalLM
 tokenizer_type: AutoTokenizer
 # Trust remote code for untrusted source
 trust_remote_code:
-# use_fast option for tokenizer loading from_pretrained, default to True
-tokenizer_use_fast:
 
 # whether you are training a 4-bit GPTQ quantized model
 gptq: true

docker/Dockerfile-base

@@ -77,7 +77,7 @@ FROM base-builder
 RUN python3 -m pip uninstall -y apex
 RUN git clone https://github.com/NVIDIA/apex
 #  `MAX_JOBS=1` disables parallel building to avoid cpu memory OOM when building image on GitHub Action (standard) runners
-RUN cd apex && MAX_JOBS=1 python3 -m pip install -v --disable-pip-version-check --no-cache-dir --no-build-isolation --config-settings "--build-option=--cpp_ext" --config-settings "--build-option=--cuda_ext" ./
+RUN cd apex && MAX_JOBS=1 python3 -m pip install --global-option="--cpp_ext" --global-option="--cuda_ext" --no-cache -v --disable-pip-version-check .
 
 RUN mkdir -p /workspace/builds
 COPY --from=bnb-builder /workspace/bitsandbytes /workspace/builds/bitsandbytes

src/axolotl/datasets.py

@@ -126,7 +126,6 @@ class ConstantLengthDataset(IterableDataset):
                     buffer_len = 0
 
                 if example:
-                    # FIXME
                     # just going to drop data points that are too long
                     if len(example["input_ids"]) <= self.seq_length:
                         input_ids = example["input_ids"]

src/axolotl/prompt_strategies/alpaca_chat.py

@@ -6,7 +6,7 @@ from axolotl.prompt_tokenizers import (
     AlpacaPromptTokenizingStrategy,
     InstructionPromptTokenizingStrategy,
 )
-from axolotl.prompters import AlpacaPrompter, PromptStyle, UnpromptedPrompter
+from axolotl.prompters import AlpacaPrompter, PromptStyle
 
 
 def load(tokenizer, cfg):
@@ -45,10 +45,8 @@ class NoSystemPrompter(AlpacaPrompter):
     Null Prompter with no system prompts
     """
 
-    system_prompt = ""
-    system_no_input_prompt = ""
-    turn_format = "{instruction} {input} "
-    turn_no_input_format = "{instruction} "
+    prompt_input = "{instruction} {input} "
+    prompt_no_input = "{instruction} "
 
     def __init__(self):  # pylint: disable=super-init-not-called
         pass
@@ -105,12 +103,3 @@ def load_camel_ai(tokenizer, cfg):
         cfg.train_on_inputs,
         cfg.sequence_len,
     )
-
-
-def load_no_prompt(tokenizer, cfg):
-    return AlpacaPromptTokenizingStrategy(
-        UnpromptedPrompter(PromptStyle.CHAT.value),
-        tokenizer,
-        cfg.train_on_inputs,
-        cfg.sequence_len,
-    )

src/axolotl/prompt_strategies/alpaca_instruct.py

@@ -1,7 +1,7 @@
 """Module loading the AlpacaInstructPromptTokenizingStrategy class"""
 
 from axolotl.prompt_tokenizers import AlpacaPromptTokenizingStrategy
-from axolotl.prompters import AlpacaPrompter, PromptStyle, UnpromptedPrompter
+from axolotl.prompters import AlpacaPrompter, PromptStyle
 
 
 def load(tokenizer, cfg):
@@ -11,12 +11,3 @@ def load(tokenizer, cfg):
         cfg.train_on_inputs,
         cfg.sequence_len,
     )
-
-
-def load_no_prompt(tokenizer, cfg):
-    return AlpacaPromptTokenizingStrategy(
-        UnpromptedPrompter(PromptStyle.INSTRUCT.value),
-        tokenizer,
-        cfg.train_on_inputs,
-        cfg.sequence_len,
-    )

src/axolotl/prompt_strategies/alpaca_w_system.py

@@ -1,84 +0,0 @@
-"""
-Prompt strategies loader for alpaca instruction datasets with system prompts
-"""
-from typing import Generator, Tuple, Union
-
-from axolotl.prompt_tokenizers import PromptTokenizingStrategy
-from axolotl.prompters import AlpacaPrompter, PromptStyle
-
-
-class InstructionWSystemPromptTokenizingStrategy(PromptTokenizingStrategy):
-    """
-    Tokenizing strategy for instruction-based prompts.
-    """
-
-    def parse_instruction_fields(self, prompt) -> Tuple[str, str, str, str]:
-        return (
-            prompt["instruction"],
-            prompt["input"] if "input" in prompt else "",
-            prompt["output"],
-            prompt["system"],
-        )
-
-    def tokenize_prompt(self, prompt):
-        # pylint: disable=duplicate-code
-        (
-            instruction,
-            input,  # pylint: disable=redefined-builtin
-            response,
-            system,
-        ) = self.parse_instruction_fields(prompt)
-        user_prompt = next(
-            iter(
-                self.prompter.build_prompt_w_system(
-                    system,
-                    instruction,
-                    input,
-                )
-            )
-        )
-        tokenized_prompt = self._tokenize(user_prompt, add_eos_token=False)
-        if not self.train_on_inputs:
-            user_prompt_len = len(tokenized_prompt["input_ids"])
-            # TODO this could be sped up using numpy array slicing
-            tokenized_prompt["labels"] = [-100] * user_prompt_len
-        tokenized_res_prompt = self._tokenize(
-            response, strip_bos_token=True, add_eos_token=True
-        )
-        tokenized_prompt["input_ids"] += tokenized_res_prompt["input_ids"]
-        tokenized_prompt["attention_mask"] += tokenized_res_prompt["attention_mask"]
-        tokenized_prompt["labels"] += tokenized_res_prompt["input_ids"]
-
-        return tokenized_prompt
-
-
-class SystemDataPrompter(AlpacaPrompter):
-    """
-    Alpaca Style Prompter that uses system prompts from the dataset
-    """
-
-    def build_prompt_w_system(
-        self,
-        system: str,
-        instruction: str,
-        input: Union[None, str] = None,  # pylint: disable=redefined-builtin
-        output: Union[None, str] = None,
-    ) -> Generator[str, None, None]:
-        # returns the full prompt from instruction and optional input
-        # if a label (=response, =output) is provided, it's also appended.
-        if input:
-            res = system + self.turn_format.format(instruction=instruction, input=input)
-        else:
-            res = system + self.turn_no_input_format.format(instruction=instruction)
-        if output:
-            res = f"{res}{output}"
-        yield res
-
-
-def load(tokenizer, cfg):
-    return InstructionWSystemPromptTokenizingStrategy(
-        SystemDataPrompter(PromptStyle.CHAT.value),
-        tokenizer,
-        cfg.train_on_inputs,
-        cfg.sequence_len,
-    )

src/axolotl/prompt_tokenizers.py

@@ -73,8 +73,17 @@ class PromptTokenizingStrategy(abc.ABC):
         ):
             result["input_ids"].append(self.tokenizer.eos_token_id)
             result["attention_mask"].append(1)
+        elif (  # some tokenizers automatically add an eos token, let's remove it
+            not add_eos_token and result["input_ids"][-1] == self.tokenizer.eos_token_id
+        ):
+            result["input_ids"] = result["input_ids"][:-1]
+            result["attention_mask"] = result["attention_mask"][:-1]
 
-        if result["input_ids"][0] == self.tokenizer.bos_token_id and strip_bos_token:
+        if (
+            self.tokenizer.bos_token_id
+            and result["input_ids"][0] == self.tokenizer.bos_token_id
+            and strip_bos_token
+        ):
             result["input_ids"] = result["input_ids"][1:]
             result["attention_mask"] = result["attention_mask"][1:]
 
@@ -87,9 +96,7 @@ class InstructionPromptTokenizingStrategy(PromptTokenizingStrategy):
     Tokenizing strategy for instruction-based prompts.
     """
 
-    def parse_instruction_fields(
-        self, prompt
-    ) -> Union[Tuple[str, str, str], Tuple[str, str, str, str]]:
+    def parse_instruction_fields(self, prompt) -> Tuple[str, str, str]:
         raise NotImplementedError
 
     def tokenize_prompt(self, prompt):
@@ -414,7 +421,11 @@ class ShareGPTPromptTokenizingStrategy(PromptTokenizingStrategy):
             result["input_ids"].append(self.tokenizer.eos_token_id)
             result["attention_mask"].append(1)
 
-        if result["input_ids"][0] == self.tokenizer.bos_token_id and strip_bos_token:
+        if (
+            self.tokenizer.bos_token_id
+            and result["input_ids"][0] == self.tokenizer.bos_token_id
+            and strip_bos_token
+        ):
             result["input_ids"] = result["input_ids"][1:]
             result["attention_mask"] = result["attention_mask"][1:]
 

src/axolotl/prompters.py

@@ -24,8 +24,6 @@ class AlpacaPrompter:
 
     system_prompt = "Below is an instruction that describes a task, paired with an input that provides further context. Write a response that appropriately completes the request.\n\n"
     system_no_input_prompt = "Below is an instruction that describes a task. Write a response that appropriately completes the request.\n\n"
-    turn_format: str
-    turn_no_input_format: str
     prompt_style: Optional[PromptStyle] = None
 
     def __init__(self, prompt_style=PromptStyle.INSTRUCT.value):
@@ -34,13 +32,23 @@ class AlpacaPrompter:
 
     def match_prompt_style(self):
         if self.prompt_style == PromptStyle.INSTRUCT.value:
-            self.turn_format = "### Instruction:\n{instruction}\n\n### Input:\n{input}\n\n### Response:\n"
-            self.turn_no_input_format = (
-                "### Instruction:\n{instruction}\n\n### Response:\n"
+            self.prompt_input = (
+                self.system_prompt
+                + "### Instruction:\n{instruction}\n\n### Input:\n{input}\n\n### Response:\n"
             )
+            self.prompt_no_input = (
+                self.system_no_input_prompt
+                + "### Instruction:\n{instruction}\n\n### Response:\n"
+            )
+            self.response_split = "### Response:"
         if self.prompt_style == PromptStyle.CHAT.value:
-            self.turn_format = "USER: {instruction}\n{input}\nASSISTANT:"
-            self.turn_no_input_format = "USER: {instruction}\nASSISTANT:"
+            self.prompt_input = (
+                self.system_prompt + "USER: {instruction}\n{input}\nASSISTANT:"
+            )
+            self.prompt_no_input = (
+                self.system_no_input_prompt + "USER: {instruction}\nASSISTANT:"
+            )
+            self.response_split = "ASSISTANT:"
 
     def build_prompt(
         self,
@@ -51,17 +59,16 @@ class AlpacaPrompter:
         # returns the full prompt from instruction and optional input
         # if a label (=response, =output) is provided, it's also appended.
         if input:
-            res = self.system_prompt + self.turn_format.format(
-                instruction=instruction, input=input
-            )
+            res = self.prompt_input.format(instruction=instruction, input=input)
         else:
-            res = self.system_no_input_prompt + self.turn_no_input_format.format(
-                instruction=instruction
-            )
+            res = self.prompt_no_input.format(instruction=instruction)
         if output:
             res = f"{res}{output}"
         yield res
 
+    def get_response(self, output: str) -> str:
+        return output.split(self.response_split)[1].strip()
+
 
 class UnpromptedPrompter(AlpacaPrompter):
     """
@@ -86,10 +93,7 @@ class MultipleChoiceExplainPrompter(AlpacaPrompter):
     """
 
     system_prompt = (
-        "Choose the answer that best answers the question. Explain your reasoning.\n"
-    )
-    system_no_input_prompt = (
-        "Choose the answer that best answers the question. Explain your reasoning.\n"
+        "Choose the answer that best answers the question. Explain your reasoning."
     )
 
 
@@ -98,12 +102,7 @@ class MultipleChoiceConcisePrompter(AlpacaPrompter):
     Prompter for multiple choice concise
     """
 
-    system_prompt = "Choose the answer that best answers the question. Be concise in your response.\n\n"
-    system_no_input_prompt = "Choose the answer that best answers the question. Be concise in your response.\n\n"
-
-    def match_prompt_style(self):
-        self.turn_format = "USER: {instruction}\n{input}\nASSISTANT:"
-        self.turn_no_input_format = "USER: {instruction}\nASSISTANT:"
+    prompt_input = "Choose the answer that best answers the question. Be concise in your response.\n\nUSER: {instruction}\n{input}\nASSISTANT:\n"
 
 
 class SummarizeTLDRPrompter(AlpacaPrompter):
@@ -111,12 +110,9 @@ class SummarizeTLDRPrompter(AlpacaPrompter):
     Prompter for summarize TLDR
     """
 
-    system_prompt = ""
-    system_no_input_prompt = ""
-
-    def match_prompt_style(self):
-        self.turn_format = "USER: Summarize the following article as a TL;DR.\n{instruction}\n{input}\nASSISTANT:"
-        self.turn_no_input_format = "USER: Summarize the following article as a TL;DR.\n{instruction}\nASSISTANT:"
+    prompt_no_input = (
+        "USER: Summarize the following article as a TL;DR.\n{instruction}\nASSISTANT:"
+    )
 
 
 class CompletionPrompter:
@@ -132,6 +128,9 @@ class CompletionPrompter:
     ) -> Generator[str, None, None]:
         yield instruction
 
+    def get_response(self, output: str) -> str:
+        return output.strip()
+
 
 class GPTeacherPrompter(AlpacaPrompter):
     """
@@ -211,6 +210,9 @@ class ReflectAlpacaPrompter:
             res = f"{res}{label}"
         yield res
 
+    def get_response(self, output: str) -> str:
+        return output.split(self.response_split)[1].strip()
+
 
 class SeparatorStyle(Enum):
     """Different separator style."""
@@ -287,6 +289,12 @@ class ShareGPTPrompter:  # pylint: disable=too-few-public-methods
             sep2=" ",
         )
 
+    # def match_prompt_style(self):
+    #     if self.prompt_style == PromptStyle.chat.value:
+    #         self.prompt_input = self.system_prompt + "USER: {instruction}\n{input}\nASSISTANT:"
+    #         self.prompt_no_input = self.system_no_input_prompt + "USER: {instruction}\nASSISTANT:"
+    #         self.response_split = "ASSISTANT:"
+
     def build_prompt(self, source) -> Generator[str, None, None]:
         # ignore the system prompt if provided
         if source[0]["from"] == "system":

src/axolotl/utils/models.py

@@ -34,20 +34,15 @@ def load_tokenizer(
     tokenizer_type,
     cfg,
 ):
-    use_fast = True  # this is the default
-    if cfg.tokenizer_use_fast is not None:
-        use_fast = cfg.tokenizer_use_fast
     if tokenizer_type:
         tokenizer = getattr(transformers, tokenizer_type).from_pretrained(
             tokenizer_config,
             trust_remote_code=cfg.trust_remote_code or False,
-            use_fast=use_fast,
         )
     else:
         tokenizer = AutoTokenizer.from_pretrained(
             tokenizer_config,
             trust_remote_code=cfg.trust_remote_code or False,
-            use_fast=use_fast,
         )
 
     logging.debug(f"EOS: {tokenizer.eos_token_id} / {tokenizer.eos_token}")

src/axolotl/utils/tokenization.py

@@ -34,5 +34,3 @@ def check_example_labels(example, tokenizer):
 
     logging.info(" ".join(colored_tokens))
     logging.info("\n\n\n")
-
-    return " ".join(colored_tokens)

src/axolotl/utils/trainer.py

@@ -124,10 +124,6 @@ def setup_trainer(cfg, train_dataset, eval_dataset, model, tokenizer):
     if cfg.max_grad_norm:
         training_arguments_kwargs["max_grad_norm"] = cfg.max_grad_norm
 
-    if cfg.push_to_hub_model_id:
-        training_arguments_kwargs["push_to_hub_model_id"] = cfg.push_to_hub_model_id
-        training_arguments_kwargs["push_to_hub"] = True
-
     training_args = transformers.TrainingArguments(
         per_device_train_batch_size=cfg.micro_batch_size,
         per_device_eval_batch_size=cfg.eval_batch_size

tests/test_prompt_tokenizers.py

@@ -7,15 +7,11 @@ from pathlib import Path
 from transformers import AutoTokenizer
 
 from axolotl.prompt_strategies.alpaca_chat import NoSystemPrompter
-from axolotl.prompt_strategies.alpaca_w_system import (
-    InstructionWSystemPromptTokenizingStrategy,
-    SystemDataPrompter,
-)
 from axolotl.prompt_tokenizers import (
     AlpacaPromptTokenizingStrategy,
     ShareGPTPromptTokenizingStrategy,
 )
-from axolotl.prompters import AlpacaPrompter, PromptStyle, ShareGPTPrompter
+from axolotl.prompters import AlpacaPrompter, ShareGPTPrompter
 
 logging.basicConfig(level="INFO")
 
@@ -100,39 +96,5 @@ class TestPromptTokenizationStrategies(unittest.TestCase):
         assert example["labels"][world_idx - 1] == -100
 
 
-class InstructionWSystemPromptTokenizingStrategyTest(unittest.TestCase):
-    """
-    Test class for prompt tokenization strategies with sys prompt from the dataset
-    """
-
-    def setUp(self) -> None:
-        # pylint: disable=duplicate-code
-        self.tokenizer = AutoTokenizer.from_pretrained("huggyllama/llama-7b")
-        self.tokenizer.add_special_tokens(
-            {
-                "bos_token": "<s>",
-                "eos_token": "</s>",
-                "unk_token": "<unk>",
-            }
-        )
-
-    def test_system_alpaca(self):
-        prompter = SystemDataPrompter(PromptStyle.CHAT.value)
-        strat = InstructionWSystemPromptTokenizingStrategy(
-            prompter,
-            self.tokenizer,
-            False,
-            2048,
-        )
-        sample = {
-            "system": "use cot",
-            "instruction": "hello!",
-            "output": "Hi! How can I help?",
-        }
-        example = strat.tokenize_prompt(sample)
-        assert example["input_ids"][0:3] == [1, 671, 20118]  # <s>use cot
-        assert example["input_ids"][3] == 11889  # USER
-
-
 if __name__ == "__main__":
     unittest.main()

tests/test_prompters.py

@@ -2,13 +2,7 @@
 
 import unittest
 
-from axolotl.prompt_strategies.alpaca_w_system import SystemDataPrompter
-from axolotl.prompters import (
-    AlpacaPrompter,
-    MultipleChoiceExplainPrompter,
-    PromptStyle,
-    UnpromptedPrompter,
-)
+from axolotl.prompters import AlpacaPrompter, PromptStyle
 
 
 class AlpacaPrompterTest(unittest.TestCase):
@@ -61,64 +55,3 @@ class AlpacaPrompterTest(unittest.TestCase):
         assert "### Response:" not in res
         assert "USER:" in res
         assert "ASSISTANT:" in res
-
-    def test_system_prompt(self):
-        prompter = SystemDataPrompter(prompt_style=PromptStyle.CHAT.value)
-        res = next(
-            prompter.build_prompt_w_system(
-                "use cot", "tell me a joke about the following", "alpacas"
-            )
-        )
-        assert "use cot" in res
-        assert res.startswith("use cot")
-        assert "### Instruction:" not in res
-        assert "### Input:" not in res
-        assert "alpacas" in res
-        assert "### Response:" not in res
-        assert "USER:" in res
-        assert "ASSISTANT:" in res
-
-
-class UnpromptedPrompterTest(unittest.TestCase):
-    """
-    Test class for UnpromptedPrompter with no system prompts
-    """
-
-    def test_prompt_style_w_none(self):
-        prompter = UnpromptedPrompter(prompt_style=None)
-        res = next(prompter.build_prompt("tell me a joke"))
-        assert "### Instruction:" in res
-        assert "tell me a joke" in res
-        assert res.startswith("###")
-
-    def test_prompt_style_w_instruct(self):
-        prompter = UnpromptedPrompter(prompt_style=PromptStyle.INSTRUCT.value)
-        res = next(
-            prompter.build_prompt("tell me a joke about the following", "alpacas")
-        )
-        assert "### Instruction:" in res
-        assert "tell me a joke" in res
-        assert res.startswith("###")
-
-    def test_prompt_style_w_chat(self):
-        prompter = UnpromptedPrompter(prompt_style=PromptStyle.CHAT.value)
-        res = next(
-            prompter.build_prompt("tell me a joke about the following", "alpacas")
-        )
-        assert "USER:" in res
-        assert "tell me a joke" in res
-        assert res.startswith("USER:")
-
-
-class MultipleChoiceExplainPrompterTest(unittest.TestCase):
-    """
-    Test class for MultipleChoiceExplainPrompter
-    """
-
-    def test_prompt_style_w_chat(self):
-        prompter = MultipleChoiceExplainPrompter(prompt_style=PromptStyle.CHAT.value)
-        res = next(prompter.build_prompt("choose one", "- A\n- B\n- C", "C"))
-        assert "USER:" in res
-        assert "choose one" in res
-        assert "Choose the answer that best answers the question." in res
-        assert "- A\n- B\n- C" in res

tests/test_tokenizers.py

@@ -1,31 +0,0 @@
-"""
-Test cases for the tokenizer loading
-"""
-import unittest
-
-from axolotl.utils.dict import DictDefault
-from axolotl.utils.models import load_tokenizer
-
-
-class TestTokenizers(unittest.TestCase):
-    """
-    test class for the load_tokenizer fn
-    """
-
-    def test_default_use_fast(self):
-        cfg = DictDefault({})
-        tokenizer = load_tokenizer("huggyllama/llama-7b", None, cfg)
-        assert "Fast" in tokenizer.__class__.__name__
-
-    def test_dont_use_fast(self):
-        cfg = DictDefault(
-            {
-                "tokenizer_use_fast": False,
-            }
-        )
-        tokenizer = load_tokenizer("huggyllama/llama-7b", None, cfg)
-        assert "Fast" not in tokenizer.__class__.__name__
-
-
-if __name__ == "__main__":
-    unittest.main()