How to make small-llama2 model work (Returns nonesense)

[CycloneRing]

Hi, I'm testing simple chat sample to use small-llama2.Q4_0.gguf
I tested tinyllama-1.1b-chat-v0.3.Q4_0.gguf and it works fine with this small change :

    // initialize the context
    llama_context_params ctx_params = llama_context_default_params();
    ctx_params.n_ctx = n_ctx;
    ctx_params.n_batch = n_ctx;
    ctx_params.type_k = GGML_TYPE_Q4_0;
    ctx_params.type_v = GGML_TYPE_Q4_0;
    ctx_params.flash_attn = true;

but it doesn't work with small llama2, Is there any specific thing I need to do?