Clean up conflict markers; finalize RL data split implementation; fix config schema conflicts; add truncation+post-filter behavior and alias handling

src/axolotl/utils/data/rl.py

@@ -122,18 +122,6 @@ def _map_dataset(
     return dataset
 
 
-<<<<<<< HEAD
-def drop_long_rl_seq(
-    sample,
-    rl,
-    tokenizer,
-    sequence_len,
-    handling="drop",  # Use the default handling mode
-):
-    result = None
-
-    if rl in (RLType.DPO, RLType.IPO, RLType.ORPO, RLType.SIMPO):
-=======
 def _drop_long_sequences(
     sample: dict[str, Any], rl: RLType, tokenizer: Any, sequence_len: int
 ) -> bool:
@@ -152,7 +140,6 @@ def _drop_long_sequences(
         ValueError: If required keys are missing or RL type is unknown.
     """
     if rl in {RLType.DPO, RLType.IPO, RLType.ORPO, RLType.SIMPO}:
->>>>>>> origin/main
         if not (
             sample.get("prompt") and sample.get("chosen") and sample.get("rejected")
         ):
@@ -169,6 +156,7 @@ def _drop_long_sequences(
         len_rejected = len(tokenizer(rejected, add_special_tokens=False)["input_ids"])
 
         # Truncate first, then drop if still invalid (although truncate should handle it)
+        handling = sample.get("sequence_len_overflow_handling", "drop")
         if handling == "truncate":
             # If both sequences fit, return sample unchanged
             if (len_prompt + len_chosen) <= sequence_len and (
@@ -225,6 +213,7 @@ def _drop_long_sequences(
         )
 
         # Truncate first
+        handling = sample.get("sequence_len_overflow_handling", "drop")
         if handling == "truncate":
             # If sequence fits, return sample unchanged
             if (len_prompt + len_completion) <= sequence_len:
@@ -255,17 +244,14 @@ def _drop_long_sequences(
             result = (len_prompt + len_completion) <= sequence_len
 
     elif rl == RLType.GRPO:
-        # GRPO doesn't involve sequence length checks in the same way?
-        # The original code returned True for drop. What should it return for truncate?
-        # Let's assume for now it always passes.
-        result = sample if handling == "truncate" else True
+        # For GRPO always keep
+        result = True
     else:
         raise ValueError("Unknown RL type")
 
     return result
 
 
-<<<<<<< HEAD
 def load_prepare_preference_datasets(cfg):
     def _is_rl_seq_within_sequence_len(sample, rl, tokenizer, sequence_len):
         """
@@ -310,10 +296,8 @@ def load_prepare_preference_datasets(cfg):
                 config_dataset, use_auth_token, streaming=False
             )
             split_datasets.append(ds)
-=======
 def _load_split(cfg: DictDefault, split: Literal["train", "test"]) -> Dataset:
     """Load and process dataset split for RL training.
->>>>>>> origin/main
 
     Args:
         cfg: Configuration object containing dataset settings.
@@ -325,7 +309,6 @@ def _load_split(cfg: DictDefault, split: Literal["train", "test"]) -> Dataset:
     datasets_configs = cfg.datasets if split == "train" else cfg.test_datasets
     split_datasets: list[Dataset | DatasetDict] = []
 
-<<<<<<< HEAD
                 map_kwargs = {}
                 if isinstance(ds_transform_fn, tuple):
                     ds_transform_fn, map_kwargs = ds_transform_fn
@@ -461,11 +444,9 @@ def _load_split(cfg: DictDefault, split: Literal["train", "test"]) -> Dataset:
     if cfg.dataset_exact_deduplication:
         train_dataset, eval_dataset, _ = deduplicate_and_log_datasets(
             train_dataset=train_dataset, eval_dataset=eval_dataset
-=======
     for dataset_config in datasets_with_name_generator(datasets_configs):
         dataset: Dataset | DatasetDict = load_dataset_with_config(
             dataset_config, cfg.hf_use_auth_token, streaming=False
->>>>>>> origin/main
         )
         split_datasets.append(dataset)