diff --git a/__init__.py b/__init__.py
new file mode 100644
index 0000000..c598173
--- /dev/null
+++ b/__init__.py
@@ -0,0 +1 @@
+__version__ = "0.9.3"
diff --git a/gritlm/gritlm.py b/gritlm.py
similarity index 100%
rename from gritlm/gritlm.py
rename to gritlm.py
diff --git a/gritlm/__init__.py b/gritlm/__init__.py
deleted file mode 100644
index 5dba751..0000000
--- a/gritlm/__init__.py
+++ /dev/null
@@ -1,3 +0,0 @@
-__version__ = "0.9.2"
-
-from gritlm import GritLM
\ No newline at end of file
diff --git a/setup.py b/setup.py
index 1f01bea..17f2cdc 100644
--- a/setup.py
+++ b/setup.py
@@ -43,7 +43,7 @@
 
 setup(
     name='gritlm',
-    version='0.9.2',
+    version='0.9.3',
     description='GritLM',
     long_description=readme,
     long_description_content_type="text/markdown",
diff --git a/gritlm/training/GradCache/.gitignore b/training/GradCache/.gitignore
similarity index 100%
rename from gritlm/training/GradCache/.gitignore
rename to training/GradCache/.gitignore
diff --git a/gritlm/training/GradCache/LICENSE b/training/GradCache/LICENSE
similarity index 100%
rename from gritlm/training/GradCache/LICENSE
rename to training/GradCache/LICENSE
diff --git a/gritlm/training/GradCache/README.md b/training/GradCache/README.md
similarity index 100%
rename from gritlm/training/GradCache/README.md
rename to training/GradCache/README.md
diff --git a/gritlm/training/GradCache/setup.py b/training/GradCache/setup.py
similarity index 100%
rename from gritlm/training/GradCache/setup.py
rename to training/GradCache/setup.py
diff --git a/gritlm/training/GradCache/src/grad_cache/__init__.py b/training/GradCache/src/grad_cache/__init__.py
similarity index 100%
rename from gritlm/training/GradCache/src/grad_cache/__init__.py
rename to training/GradCache/src/grad_cache/__init__.py
diff --git a/gritlm/training/GradCache/src/grad_cache/cachex/__init__.py b/training/GradCache/src/grad_cache/cachex/__init__.py
similarity index 100%
rename from gritlm/training/GradCache/src/grad_cache/cachex/__init__.py
rename to training/GradCache/src/grad_cache/cachex/__init__.py
diff --git a/gritlm/training/GradCache/src/grad_cache/cachex/functional.py b/training/GradCache/src/grad_cache/cachex/functional.py
similarity index 100%
rename from gritlm/training/GradCache/src/grad_cache/cachex/functional.py
rename to training/GradCache/src/grad_cache/cachex/functional.py
diff --git a/gritlm/training/GradCache/src/grad_cache/cachex/training.py b/training/GradCache/src/grad_cache/cachex/training.py
similarity index 100%
rename from gritlm/training/GradCache/src/grad_cache/cachex/training.py
rename to training/GradCache/src/grad_cache/cachex/training.py
diff --git a/gritlm/training/GradCache/src/grad_cache/cachex/tree_utils.py b/training/GradCache/src/grad_cache/cachex/tree_utils.py
similarity index 100%
rename from gritlm/training/GradCache/src/grad_cache/cachex/tree_utils.py
rename to training/GradCache/src/grad_cache/cachex/tree_utils.py
diff --git a/gritlm/training/GradCache/src/grad_cache/context_managers.py b/training/GradCache/src/grad_cache/context_managers.py
similarity index 100%
rename from gritlm/training/GradCache/src/grad_cache/context_managers.py
rename to training/GradCache/src/grad_cache/context_managers.py
diff --git a/gritlm/training/GradCache/src/grad_cache/functional.py b/training/GradCache/src/grad_cache/functional.py
similarity index 100%
rename from gritlm/training/GradCache/src/grad_cache/functional.py
rename to training/GradCache/src/grad_cache/functional.py
diff --git a/gritlm/training/GradCache/src/grad_cache/grad_cache.py b/training/GradCache/src/grad_cache/grad_cache.py
similarity index 100%
rename from gritlm/training/GradCache/src/grad_cache/grad_cache.py
rename to training/GradCache/src/grad_cache/grad_cache.py
diff --git a/gritlm/training/GradCache/src/grad_cache/loss.py b/training/GradCache/src/grad_cache/loss.py
similarity index 100%
rename from gritlm/training/GradCache/src/grad_cache/loss.py
rename to training/GradCache/src/grad_cache/loss.py
diff --git a/gritlm/training/arguments.py b/training/arguments.py
similarity index 100%
rename from gritlm/training/arguments.py
rename to training/arguments.py
diff --git a/gritlm/training/data.py b/training/data.py
similarity index 100%
rename from gritlm/training/data.py
rename to training/data.py
diff --git a/gritlm/training/gradcache_trainer.py b/training/gradcache_trainer.py
similarity index 100%
rename from gritlm/training/gradcache_trainer.py
rename to training/gradcache_trainer.py
diff --git a/gritlm/training/model.py b/training/model.py
similarity index 100%
rename from gritlm/training/model.py
rename to training/model.py
diff --git a/gritlm/training/run.py b/training/run.py
similarity index 100%
rename from gritlm/training/run.py
rename to training/run.py
diff --git a/gritlm/training/toy_data/toy_data_embedding.jsonl b/training/toy_data/toy_data_embedding.jsonl
similarity index 100%
rename from gritlm/training/toy_data/toy_data_embedding.jsonl
rename to training/toy_data/toy_data_embedding.jsonl
diff --git a/gritlm/training/toy_data/toy_data_generative.jsonl b/training/toy_data/toy_data_generative.jsonl
similarity index 100%
rename from gritlm/training/toy_data/toy_data_generative.jsonl
rename to training/toy_data/toy_data_generative.jsonl
diff --git a/gritlm/training/toy_data_instruct/toy_data_embedding.jsonl b/training/toy_data_instruct/toy_data_embedding.jsonl
similarity index 100%
rename from gritlm/training/toy_data_instruct/toy_data_embedding.jsonl
rename to training/toy_data_instruct/toy_data_embedding.jsonl
diff --git a/gritlm/training/toy_data_instruct/toy_data_generative.jsonl b/training/toy_data_instruct/toy_data_generative.jsonl
similarity index 100%
rename from gritlm/training/toy_data_instruct/toy_data_generative.jsonl
rename to training/toy_data_instruct/toy_data_generative.jsonl