fix: Fix indexing into k_l for recurrent cache with filter

Branch: HybridCache

Signed-off-by: Gabe Goodhart <[email protected]>

Author: gabe-l-hart

src/llama-kv-cache.cpp

@@ -1906,8 +1906,8 @@ llama_kv_cache_recurrent::llama_kv_cache_recurrent(
         ggml_tensor * v = ggml_new_tensor_1d(ctx, type_v, n_embd_v_gqa*kv_size);
         ggml_format_name(k, "cache_k_l%d", i);
         ggml_format_name(v, "cache_v_l%d", i);
-        k_l.push_back(k);
-        v_l.push_back(v);
+        k_l[i] = k;
+        v_l[i] = v;
     }
 
     // allocate tensors and initialize the buffers to avoid NaNs in the padding