perform flakey patched tests in individual runner

README.md

@@ -478,7 +478,7 @@ See [examples](examples) for quick start. It is recommended to duplicate and mod
 
 - model
   ```yaml
-  base_model: ./llama-7b-hf/ # local or huggingface repo
+  base_model: ./llama-7b-hf # local or huggingface repo
   ```
   Note: The code will load the right architecture.
 

cicd/cicd.sh

@@ -8,3 +8,8 @@ pytest -v --durations=10 -n8 --ignore=tests/e2e/ --ignore=tests/patched/ /worksp
 pytest -v --durations=10 -n1 --dist loadfile /workspace/axolotl/tests/e2e/patched/
 pytest -v --durations=10 -n1 --dist loadfile /workspace/axolotl/tests/e2e/integrations/
 pytest -v --durations=10 --ignore=tests/e2e/patched/ --ignore=tests/e2e/multigpu/ --ignore=tests/e2e/integrations/ /workspace/axolotl/tests/e2e/
+
+tests=$(pytest --collect-only -q tests/e2e/each)
+for t in $tests; do
+    pytest $t
+done

requirements.txt

@@ -12,7 +12,7 @@ liger-kernel==0.4.2
 
 packaging==23.2
 peft==0.14.0
-transformers>=4.46.3
+transformers==4.47.0
 tokenizers>=0.20.1
 accelerate==1.2.0
 datasets==3.1.0