tocmo0nlord/axolotl
1
0
Fork 0
Code Issues Pull Requests Actions 2 Packages Projects Releases Wiki Activity

fix: transformers deprecate load_in_Xbit in model_kwargs (#3205)

Browse Source 
* fix: transformers deprecate load_in_Xbit in model_kwargs

* fix: test to read from quantization_config kwarg

* fix: test

* fix: access

* fix: test weirdly entering incorrect config
This commit is contained in:
NanoCode012
2025-10-16 16:07:27 +07:00
committed by GitHub
parent 4cdfdfebb5
commit aa1240acd8
2 changed files with 21 additions and 23 deletions
Download Patch File Download Diff File Expand all files Collapse all files

16
src/axolotl/loaders/model.py
View File

@@ -515,9 +515,6 @@ class ModelLoader:
if self.cfg.model_quantization_config_kwargs: if self.cfg.model_quantization_config_kwargs:
mxfp4_kwargs = self.cfg.model_quantization_config_kwargs mxfp4_kwargs = self.cfg.model_quantization_config_kwargs
self.model_kwargs["quantization_config"] = Mxfp4Config(**mxfp4_kwargs) self.model_kwargs["quantization_config"] = Mxfp4Config(**mxfp4_kwargs)
else:
self.model_kwargs["load_in_8bit"] = self.cfg.load_in_8bit
self.model_kwargs["load_in_4bit"] = self.cfg.load_in_4bit
if self.cfg.gptq: if self.cfg.gptq:
if not hasattr(self.model_config, "quantization_config"): if not hasattr(self.model_config, "quantization_config"):
@@ -552,9 +549,7 @@ class ModelLoader:
self.model_kwargs["quantization_config"] = BitsAndBytesConfig( self.model_kwargs["quantization_config"] = BitsAndBytesConfig(
**self.model_config.quantization_config **self.model_config.quantization_config
) )
elif self.cfg.adapter == "qlora" and self.model_kwargs.get( elif self.cfg.adapter == "qlora" and self.cfg.load_in_4bit:
"load_in_4bit", False
):
bnb_config = { bnb_config = {
"load_in_4bit": True, "load_in_4bit": True,
"llm_int8_threshold": 6.0, "llm_int8_threshold": 6.0,
@@ -580,9 +575,7 @@ class ModelLoader:
self.model_kwargs["quantization_config"] = BitsAndBytesConfig( self.model_kwargs["quantization_config"] = BitsAndBytesConfig(
**bnb_config, **bnb_config,
) )
elif self.cfg.adapter == "lora" and self.model_kwargs.get( elif self.cfg.adapter == "lora" and self.cfg.load_in_8bit:
"load_in_8bit", False
):
bnb_config = { bnb_config = {
"load_in_8bit": True, "load_in_8bit": True,
} }
@@ -596,11 +589,6 @@ class ModelLoader:
**bnb_config, **bnb_config,
) )
# no longer needed per https://github.com/huggingface/transformers/pull/26610
if "quantization_config" in self.model_kwargs or self.cfg.gptq:
self.model_kwargs.pop("load_in_8bit", None)
self.model_kwargs.pop("load_in_4bit", None)
def _set_attention_config(self): def _set_attention_config(self):
"""Sample packing uses custom FA2 patch""" """Sample packing uses custom FA2 patch"""
if self.cfg.attn_implementation: if self.cfg.attn_implementation:

28
tests/test_loaders.py
View File

@@ -80,16 +80,26 @@ class TestModelsUtils:
hasattr(self.model_loader.model_kwargs, "load_in_8bit") hasattr(self.model_loader.model_kwargs, "load_in_8bit")
and hasattr(self.model_loader.model_kwargs, "load_in_4bit") and hasattr(self.model_loader.model_kwargs, "load_in_4bit")
) )
elif load_in_8bit and self.cfg.adapter is not None:
assert self.model_loader.model_kwargs["load_in_8bit"]
elif load_in_4bit and self.cfg.adapter is not None:
assert self.model_loader.model_kwargs["load_in_4bit"]
if (self.cfg.adapter == "qlora" and load_in_4bit) or ( if self.cfg.adapter == "qlora" and load_in_4bit:
self.cfg.adapter == "lora" and load_in_8bit assert isinstance(
): self.model_loader.model_kwargs.get("quantization_config"),
assert self.model_loader.model_kwargs.get( BitsAndBytesConfig,
"quantization_config", BitsAndBytesConfig )
assert (
self.model_loader.model_kwargs["quantization_config"]._load_in_4bit
is True
)
if self.cfg.adapter == "lora" and load_in_8bit:
assert isinstance(
self.model_loader.model_kwargs.get("quantization_config"),
BitsAndBytesConfig,
)
assert (
self.model_loader.model_kwargs["quantization_config"]._load_in_8bit
is True
) )
def test_message_property_mapping(self): def test_message_property_mapping(self):