add support for CPU and MPS

do not use distributed when not available, instead use CPU or MPS.

This entails a few changes:

--device is now a valid flag to the library since `ilab` can pass CPU, MPS, or default to cuda
when using CPU or MPS, do not initialize DS, instead put the model on the device and initialize `Adafactor` optimizer which is more efficient and than Adam based one
inside of `train` add logic for handling if torch.cuda.is_available and torch.distributed.is_initialized() we dont use distributed torch on consumer systems
the train loop needs some custom step and loss logic for a LlamaForCausalLM model, add that in
when using CPU or MPS we are always world_size == 1 and local_rank == 0

Signed-off-by: Charlie Doern <[email protected]>

Author: cdoern

src/instructlab/training/__init__.py

@@ -22,9 +22,9 @@
 
 
 # defer import of main_ds
-def run_training(torch_args: TorchrunArgs, train_args: TrainingArgs) -> None:
+def run_training(torch_args: TorchrunArgs, train_args: TrainingArgs, device: str) -> None:
     """Wrapper around the main training job that calls torchrun."""
     # Local
     from .main_ds import run_training
 
-    return run_training(torch_args=torch_args, train_args=train_args)
+    return run_training(torch_args=torch_args, train_args=train_args, device=device)