llama_model_loader: - tensor split  0:                token_embd.weight f16      [   2560, 262144,      1,      1 ]  1280.00 MiB
llama_model_loader: - tensor split  0:      per_layer_token_embd.weight f16      [  10752, 262144,      1,      1 ]  5376.00 MiB
0.02.934.812 I load_tensors: offloaded 43/43 layers to GPU
0.02.934.818 I load_tensors:   CPU_Mapped model buffer size =  6656.00 MiB
0.02.934.819 I load_tensors:        CUDA0 model buffer size =  4096.00 MiB
