Fixed GPU based vLLM path references

documentation/training_and_learning/genai/README.md

@@ -47,24 +47,23 @@ You can skip to testing the vLLM after this step
 2. Create PVC, deployment, & service in __vllm__ namespace.
 
 #### PVC: 
-`oc apply -f https://raw.githubusercontent.com/rh-aiservices-bu/llm-on-openshift/main/llm-servers/vllm/gitops/pvc.yaml`
+`oc apply -f https://raw.githubusercontent.com/rh-aiservices-bu/llm-on-openshift/main/llm-servers/vllm/gpu/gitops/pvc.yaml`
 
 #### Service: 
-`oc apply -f https://raw.githubusercontent.com/rh-aiservices-bu/llm-on-openshift/main/llm-servers/vllm/gitops/service.yaml`
+`oc apply -f https://raw.githubusercontent.com/rh-aiservices-bu/llm-on-openshift/main/llm-servers/vllm/gpu/gitops/service.yaml`
 
 #### Route: 
-`oc apply -f https://raw.githubusercontent.com/rh-aiservices-bu/llm-on-openshift/main/llm-servers/vllm/gitops/route.yaml`
+`oc apply -f https://raw.githubusercontent.com/rh-aiservices-bu/llm-on-openshift/main/llm-servers/vllm/gpu/gitops/route.yaml`
 
 #### Deployment: 
-
 For the deployment, we'll need to edit the __env variable: HUGGING_FACE_HUB_TOKEN__ with your hugging face token.
 
 We need this use this token because the model we're using, _Mistral-7B-Instruct-v0.2_, is gated. You must be authenticated to access it. Navigate to https://huggingface.co/settings/tokens and create a new token and copy it for the environment variable.
 
 ![mistralhf](./images/mistralhf_login.png "mistralhf")
 
 
-Download the __deployment.yaml__: https://github.com/rh-aiservices-bu/llm-on-openshift/blob/main/llm-servers/vllm/gitops/deployment.yaml#L52 and add your token.
+Download the __deployment.yaml__: https://github.com/rh-aiservices-bu/llm-on-openshift/blob/main/llm-servers/vllm/gpu/gitops/deployment.yaml#L52 and add your token.
 
 ![env hf token](./images/hf_hub_token.png "env hf token")
 
@@ -213,4 +212,4 @@ It then goes through asking a question which then invokes the LLM with the infor
 
 ![rag results](./images/rag_results.png "rag results")
 
-As you can see above, the answer that the LLM gives is more specific to creating a Data Connection in RHOAI because it's uses the rhoai documents that are in the vector db.
+As you can see above, the answer that the LLM gives is more specific to creating a Data Connection in RHOAI because it's uses the rhoai documents that are in the vector db.