Merge branch 'main' into data-load-refactor

Author: djsaunde

_quarto.yml

@@ -17,7 +17,9 @@ quartodoc:
         - convert
         - prompt_tokenizers
         - logging_config
-        - core.trainer_builder
+        - core.builders.base
+        - core.builders.causal
+        - core.builders.rl
         - core.training_args
         - core.chat.messages
         - core.chat.format.chatml

docs/config.qmd

@@ -514,6 +514,7 @@ output_dir: ./completed-model
 # setting to `auto` will enable torch compile when torch>=2.5.1
 torch_compile:  # Optional[Union[Literal["auto"], bool]]
 torch_compile_backend:  # Optional[str]
+torch_compile_mode:  # 'default' | 'reduce-overhead' | 'max-autotune'
 
 # Training hyperparameters
 
@@ -560,7 +561,7 @@ profiler_steps: # enable the pytorch profiler to capture the first N steps of tr
 loss_watchdog_threshold: # High loss value, indicating the learning has broken down (a good estimate is ~2 times the loss at the start of training)
 loss_watchdog_patience: # Number of high-loss steps in a row before the trainer aborts (default: 3)
 
-# Save model as safetensors (require safetensors package)
+# Save model as safetensors (require safetensors package). Default True
 save_safetensors:
 
 # Whether to mask out or include the human's prompt from the training labels

src/axolotl/core/builders/__init__.py

@@ -0,0 +1,6 @@
+"""Trainer builder classes"""
+
+from .causal import HFCausalTrainerBuilder
+from .rl import HFRLTrainerBuilder
+
+__all__ = ["HFCausalTrainerBuilder", "HFRLTrainerBuilder"]