Update README.md

README.md

@@ -16,7 +16,6 @@ Quantization
 - --qembedding 8w: embeddings use 8-bit weights
 
 ## Export
-
 ```
 optimum-cli export executorch \
   --model "google/gemma-3-4b-it" \
@@ -31,15 +30,16 @@ optimum-cli export executorch \
 ```
 
 ## Run
-
-Build the runner from the ExecuTorch repo root:
+Build the runner from the ExecuTorch repo root
 ```
 make gemma3-cpu
 ```
-Binary is located at cmake-out/examples/models/gemma3/gemma3_e2e_runner
-
+Download tokenizer
+```
 curl -L https://huggingface.co/google/gemma-3-4b-it/resolve/main/tokenizer.json -o tokenizer.json
 ```
+Run model
+```
 ./cmake-out/examples/models/gemma3/gemma3_e2e_runner \
   --model_path "model.pte" \
   --tokenizer_path "tokenizer.json" \