Use their config

Author: kirklandsign

extension/android/executorch_android/android_test_setup.sh

@@ -21,7 +21,7 @@ curl -C - -Ls "https://raw.githubusercontent.com/karpathy/llama2.c/master/tokeni
 # Create params.json file
 touch params.json
 echo '{"dim": 768, "multiple_of": 32, "n_heads": 12, "n_layers": 12, "norm_eps": 1e-05, "vocab_size": 32000}' > params.json
-python -m examples.models.llama.export_llama -c stories110M.pt -p params.json -d fp16 --model=stories110m
+python -m examples.models.llama.export_llama -c stories110M.pt -p params.json -d fp16 -n stories110m_h.pte -kv
 python -m pytorch_tokenizers.tools.llama2c.convert -t tokenizer.model -o tokenizer.bin
 
 cp stories110m_h.pte "${BASEDIR}/src/androidTest/resources/stories.pte"