Small fixes

pyproject.toml

@@ -37,7 +37,7 @@ triton = [
 ]
 
 huggingface = [
-    "unsloth_zoo>=2025.7.1",
+    "unsloth_zoo>=2025.7.2",
     "packaging",
     "tyro",
     "transformers>=4.51.3,!=4.47.0,!=4.52.0,!=4.52.1,!=4.52.2,!=4.52.3",
@@ -381,7 +381,7 @@ colab-ampere-torch220 = [
     "flash-attn>=2.6.3",
 ]
 colab-new = [
-    "unsloth_zoo>=2025.7.1",
+    "unsloth_zoo>=2025.7.2",
     "packaging",
     "tyro",
     "transformers>=4.51.3,!=4.47.0,!=4.52.0,!=4.52.1,!=4.52.2,!=4.52.3",

unsloth/__init__.py

@@ -79,7 +79,7 @@ def get_device_count():
     elif DEVICE_TYPE == "xpu":
         return torch.xpu.device_count()
     else:
-        return 0
+        return 1
 pass
 
 DEVICE_COUNT : int = get_device_count()

unsloth/models/_utils.py

@@ -12,7 +12,7 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
-__version__ = "2025.7.1"
+__version__ = "2025.7.2"
 
 __all__ = [
     "SUPPORTS_BFLOAT16",

unsloth/models/llama.py

@@ -796,7 +796,7 @@ def LlamaModel_fast_forward(
     # Ignore attention_mask
     if attention_mask is None:
         padding_mask = None
-    elif self.training and os.environ.get("UNSLOTH_KEEP_PADDING", "0") != '1':
+    elif self.training:
         attention_mask = None
         padding_mask = None
     else: