instruct model setup

2025-08-28 17:57:59 +00:00
parent 77c563f358
commit d49b4ff2d5
55 changed files with 27760 additions and 326 deletions
@@ -1,7 +1,7 @@
 """
-2025.8.4
-2025.8.5
-4.55.1
+2025.8.9
+2025.8.10
+4.55.4
 0.21.0
 __UNSLOTH_VERSIONING__
 """
@@ -10,7 +10,7 @@ import torch
 import torch.nn as nn
 from torch.nn import functional as F
 from typing import Any, List, Optional, Tuple, Union, Dict, Set, Callable
-from trl.trainer.sft_trainer import (Any, AutoModelForCausalLM, AutoTokenizer, BaseImageProcessor, Callable, DataCollator, DataCollatorForLanguageModeling, Dataset, EvalPrediction, FeatureExtractionMixin, IterableDataset, Optional, Path, PeftConfig, PeftModel, PreTrainedModel, PreTrainedTokenizerBase, ProcessorMixin, SFTConfig, SFTTrainer, Trainer, TrainerCallback, TrainingArguments, Union, clone_chat_template, contextlib, dataclass, dataclasses, defaultdict, generate_model_card, get_act_offloading_ctx_manager, get_comet_experiment_url, get_peft_model, is_conversational, is_peft_available, is_wandb_available, nn, os, pad, peft, peft_module_casting_to_bf16, prepare_model_for_kbit_training, torch, version, warnings, Callable, DataCollator, DataCollatorForLanguageModeling, Dataset, IterableDataset, Optional, Union, os, pad, Optional, PeftModel, PreTrainedModel, Trainer, is_peft_available, os, peft, torch, os)
+from trl.trainer.sft_trainer import (Any, AutoModelForCausalLM, AutoTokenizer, BaseImageProcessor, Callable, DataCollator, DataCollatorForLanguageModeling, Dataset, EvalPrediction, FeatureExtractionMixin, IterableDataset, Optional, Path, PeftConfig, PeftModel, PreTrainedModel, PreTrainedTokenizerBase, ProcessorMixin, SFTConfig, SFTTrainer, Trainer, TrainerCallback, TrainingArguments, Union, clone_chat_template, contextlib, dataclass, dataclasses, defaultdict, generate_model_card, get_act_offloading_ctx_manager, get_comet_experiment_url, get_peft_model, is_conversational, is_peft_available, is_wandb_available, nn, os, pack_dataset, pad, peft, peft_module_casting_to_bf16, prepare_model_for_kbit_training, torch, version, warnings, Callable, DataCollator, DataCollatorForLanguageModeling, Dataset, IterableDataset, Optional, Union, os, pack_dataset, pad, Optional, PeftModel, PreTrainedModel, Trainer, is_peft_available, os, peft, torch, os)


 import os
@@ -132,6 +132,10 @@ class UnslothSFTConfig(SFTConfig):
        default = -1,
        metadata = {'help': 'Chunk size to reduce memory usage. -1 is most efficient.'},
    )
+    max_seq_length : Optional[int] = field(
+        default = None,
+        metadata = {'help': 'Maximum sequence length to truncate to.'},
+    )
    def __init__(
        self,
        output_dir = None,
@@ -280,6 +284,7 @@ class UnslothSFTConfig(SFTConfig):
        activation_offloading = False,
        vllm_sampling_params = None,
        unsloth_num_chunks = -1,
+        max_seq_length = None,
        **kwargs,
    ):
        if learning_rate < 1e-7: raise FloatingPointError(f'Unsloth: Your learning rate of `{learning_rate}` is too small and less than 1e-7! Consider increasing it, otherwise gradient updates will be close to 0!')
@@ -289,7 +294,13 @@ class UnslothSFTConfig(SFTConfig):
            save_strategy = 'no'
        if dataset_num_proc is None:
            from multiprocessing import cpu_count
-            dataset_num_proc = min(cpu_count()*2, 2)
+            dataset_num_proc = max(cpu_count()+4, 2)
+        if os.environ.get('UNSLOTH_ENABLE_FLEX_ATTENTION', '0') == '1':
+            from unsloth_zoo.flex_attention import HAS_FLEX_ATTENTION
+            if HAS_FLEX_ATTENTION and pad_to_multiple_of is None:
+                from unsloth_zoo.flex_attention import FLEX_ATTENTION_BLOCK_SIZE
+                pad_to_multiple_of = FLEX_ATTENTION_BLOCK_SIZE
+        
        
        super().__init__(
            output_dir = output_dir,
@@ -438,6 +449,7 @@ class UnslothSFTConfig(SFTConfig):
            activation_offloading = activation_offloading,**kwargs)
        self.vllm_sampling_params = vllm_sampling_params
        self.unsloth_num_chunks = unsloth_num_chunks
+        self.max_seq_length = max_seq_length
 pass

 class _UnslothSFTTrainer(Trainer):
@@ -868,7 +880,11 @@ class _UnslothSFTTrainer(Trainer):
            pass
    
            if not isinstance(dataset, IterableDataset):
-                map_kwargs["num_proc"] = getattr(args, "dataset_num_proc", 2)
+                dataset_num_proc = getattr(args, "dataset_num_proc", None)
+                if dataset_num_proc is None:
+                    from multiprocessing import cpu_count
+                    dataset_num_proc = max(cpu_count()+4, 2)
+                map_kwargs["num_proc"] = dataset_num_proc
            else:
                map_kwargs["batch_size"] = dataset._ex_iterable.batch_size
                
@@ -882,18 +898,22 @@ class _UnslothSFTTrainer(Trainer):
            pass
        pass
        if packing:
-            print("Unsloth: Hugging Face's packing is currently buggy - we're disabling it for now!")
-            return dataset
+            # Try using new packing which works in TRL
+            try:
+                pack_dataset
+            except:
+                print("Unsloth: Hugging Face's packing is currently buggy - we're disabling it for now!")
+                return dataset
    
            if max_seq_length == 0:
                raise ValueError("When packing is enabled, `max_seq_length` can't be `None`.")
    
            if use_desc: map_kwargs["desc"] = f"Unsloth: Packing {dataset_name} dataset"
-            dataset = dataset.select_columns(used_column_names).map(
-                pack_examples,
-                batched = True,
-                fn_kwargs = {"seq_length": max_seq_length,},
-                **map_kwargs,
+            dataset = pack_dataset(
+                dataset.select_columns(used_column_names),
+                max_seq_length,
+                getattr(args, "packing_strategy", "bfd"),
+                map_kwargs,
            )
        pass
        return dataset
@@ -1101,7 +1121,7 @@ class UnslothSFTTrainer(_UnslothSFTTrainer):
            print('Unsloth: Switching to float32 training since model cannot work with float16')
            force_float32 = True
        mixed_precision_dtype = os.environ.get('UNSLOTH_MIXED_PRECISION', 'float32')
-        dtype = getattr(model.config, 'torch_dtype', None)
+        dtype = getattr(model.config, 'dtype', None) or getattr(model.config, 'torch_dtype', None)
        if dtype is None: dtype = model.get_input_embeddings().dtype
        from unsloth_zoo.utils import _get_dtype
        dtype = _get_dtype(dtype)
@@ -1166,9 +1186,7 @@ class UnslothSFTTrainer(_UnslothSFTTrainer):
                    max_length = args.max_length
            else:
                model_max_length = getattr(model, 'max_seq_length', None)
-                # print(model_max_length, 'mml1')
                if model_max_length is None: model_max_length = getattr(model, 'max_length', None)
-                # print(model_max_length, 'mml2')
                if model_max_length is not None:
                    args.max_length = model_max_length
                    max_length = args.max_length
@@ -1189,9 +1207,17 @@ class UnslothSFTTrainer(_UnslothSFTTrainer):
        from unsloth_zoo.vision_utils import UnslothVisionDataCollator
        if not isinstance(data_collator, UnslothVisionDataCollator):
            if isinstance(data_collator, DataCollatorForSeq2Seq) and 'labels' not in train_dataset.column_names:
-                data_collator = TransformersDataCollatorForLanguageModeling(__tokenizer, mlm = False, mlm_probability = 0.0)
+                data_collator = TransformersDataCollatorForLanguageModeling(
+                    __tokenizer,
+                    mlm = False,
+                    mlm_probability = 0.0,
+                    pad_to_multiple_of = getattr(args, 'pad_to_multiple_of', None),
+                )
            elif isinstance(data_collator, TransformersDataCollatorForLanguageModeling) and 'labels' in train_dataset.column_names:
-                data_collator = DataCollatorForSeq2Seq(__tokenizer)
+                data_collator = DataCollatorForSeq2Seq(
+                    __tokenizer,
+                    pad_to_multiple_of = getattr(args, 'pad_to_multiple_of', None),
+                )
        else:
            if hasattr(args, 'remove_unused_columns'): args.remove_unused_columns = False
            if hasattr(args, 'dataset_text_field'): args.dataset_text_field = ''
@@ -1199,9 +1225,17 @@ class UnslothSFTTrainer(_UnslothSFTTrainer):
        if not isinstance(data_collator, UnslothVisionDataCollator):
            if not hasattr(__tokenizer, 'pad') and hasattr(__tokenizer, 'tokenizer'):
                if isinstance(data_collator, DataCollatorForSeq2Seq):
-                    data_collator = DataCollatorForSeq2Seq(__tokenizer.tokenizer)
+                    data_collator = DataCollatorForSeq2Seq(
+                        __tokenizer.tokenizer,
+                        pad_to_multiple_of = getattr(args, 'pad_to_multiple_of', None),
+                    )
                else:
-                    data_collator = TransformersDataCollatorForLanguageModeling(__tokenizer.tokenizer, mlm = False, mlm_probability = 0.0)
+                    data_collator = TransformersDataCollatorForLanguageModeling(
+                        __tokenizer.tokenizer,
+                        mlm = False,
+                        mlm_probability = 0.0,
+                        pad_to_multiple_of = getattr(args, 'pad_to_multiple_of', None),
+                    )
        other_metrics = []
        
        from unsloth_zoo.logging_utils import PatchRLStatistics