{ "metadata": { "total_size": 55238062080 }, "weight_map": { "lm_head.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.0.attn.k_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.0.attn.o_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.0.attn.q_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.0.attn.v_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.0.input_proj.bias": "pytorch_model-00006-of-00006.bin", "model.added_attn.0.input_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.0.output_proj.bias": "pytorch_model-00006-of-00006.bin", "model.added_attn.0.output_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.1.attn.k_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.1.attn.o_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.1.attn.q_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.1.attn.v_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.1.input_proj.bias": "pytorch_model-00006-of-00006.bin", "model.added_attn.1.input_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.1.output_proj.bias": "pytorch_model-00006-of-00006.bin", "model.added_attn.1.output_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.10.attn.k_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.10.attn.o_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.10.attn.q_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.10.attn.v_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.10.input_proj.bias": "pytorch_model-00006-of-00006.bin", "model.added_attn.10.input_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.10.output_proj.bias": "pytorch_model-00006-of-00006.bin", "model.added_attn.10.output_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.11.attn.k_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.11.attn.o_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.11.attn.q_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.11.attn.v_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.11.input_proj.bias": "pytorch_model-00006-of-00006.bin", "model.added_attn.11.input_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.11.output_proj.bias": "pytorch_model-00006-of-00006.bin", "model.added_attn.11.output_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.12.attn.k_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.12.attn.o_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.12.attn.q_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.12.attn.v_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.12.input_proj.bias": "pytorch_model-00006-of-00006.bin", "model.added_attn.12.input_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.12.output_proj.bias": "pytorch_model-00006-of-00006.bin", "model.added_attn.12.output_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.13.attn.k_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.13.attn.o_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.13.attn.q_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.13.attn.v_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.13.input_proj.bias": "pytorch_model-00006-of-00006.bin", "model.added_attn.13.input_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.13.output_proj.bias": "pytorch_model-00006-of-00006.bin", "model.added_attn.13.output_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.14.attn.k_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.14.attn.o_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.14.attn.q_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.14.attn.v_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.14.input_proj.bias": "pytorch_model-00006-of-00006.bin", "model.added_attn.14.input_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.14.output_proj.bias": "pytorch_model-00006-of-00006.bin", "model.added_attn.14.output_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.15.attn.k_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.15.attn.o_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.15.attn.q_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.15.attn.v_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.15.input_proj.bias": "pytorch_model-00006-of-00006.bin", "model.added_attn.15.input_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.15.output_proj.bias": "pytorch_model-00006-of-00006.bin", "model.added_attn.15.output_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.16.attn.k_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.16.attn.o_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.16.attn.q_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.16.attn.v_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.16.input_proj.bias": "pytorch_model-00006-of-00006.bin", "model.added_attn.16.input_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.16.output_proj.bias": "pytorch_model-00006-of-00006.bin", "model.added_attn.16.output_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.17.attn.k_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.17.attn.o_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.17.attn.q_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.17.attn.v_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.17.input_proj.bias": "pytorch_model-00006-of-00006.bin", "model.added_attn.17.input_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.17.output_proj.bias": "pytorch_model-00006-of-00006.bin", "model.added_attn.17.output_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.18.attn.k_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.18.attn.o_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.18.attn.q_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.18.attn.v_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.18.input_proj.bias": "pytorch_model-00006-of-00006.bin", "model.added_attn.18.input_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.18.output_proj.bias": "pytorch_model-00006-of-00006.bin", "model.added_attn.18.output_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.19.attn.k_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.19.attn.o_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.19.attn.q_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.19.attn.v_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.19.input_proj.bias": "pytorch_model-00006-of-00006.bin", "model.added_attn.19.input_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.19.output_proj.bias": "pytorch_model-00006-of-00006.bin", "model.added_attn.19.output_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.2.attn.k_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.2.attn.o_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.2.attn.q_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.2.attn.v_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.2.input_proj.bias": "pytorch_model-00006-of-00006.bin", "model.added_attn.2.input_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.2.output_proj.bias": "pytorch_model-00006-of-00006.bin", "model.added_attn.2.output_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.20.attn.k_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.20.attn.o_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.20.attn.q_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.20.attn.v_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.20.input_proj.bias": "pytorch_model-00006-of-00006.bin", "model.added_attn.20.input_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.20.output_proj.bias": "pytorch_model-00006-of-00006.bin", "model.added_attn.20.output_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.21.attn.k_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.21.attn.o_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.21.attn.q_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.21.attn.v_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.21.input_proj.bias": "pytorch_model-00006-of-00006.bin", "model.added_attn.21.input_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.21.output_proj.bias": "pytorch_model-00006-of-00006.bin", "model.added_attn.21.output_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.22.attn.k_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.22.attn.o_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.22.attn.q_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.22.attn.v_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.22.input_proj.bias": "pytorch_model-00006-of-00006.bin", "model.added_attn.22.input_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.22.output_proj.bias": "pytorch_model-00006-of-00006.bin", "model.added_attn.22.output_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.23.attn.k_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.23.attn.o_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.23.attn.q_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.23.attn.v_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.23.input_proj.bias": "pytorch_model-00006-of-00006.bin", "model.added_attn.23.input_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.23.output_proj.bias": "pytorch_model-00006-of-00006.bin", "model.added_attn.23.output_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.24.attn.k_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.24.attn.o_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.24.attn.q_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.24.attn.v_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.24.input_proj.bias": "pytorch_model-00006-of-00006.bin", "model.added_attn.24.input_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.24.output_proj.bias": "pytorch_model-00006-of-00006.bin", "model.added_attn.24.output_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.25.attn.k_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.25.attn.o_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.25.attn.q_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.25.attn.v_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.25.input_proj.bias": "pytorch_model-00006-of-00006.bin", "model.added_attn.25.input_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.25.output_proj.bias": "pytorch_model-00006-of-00006.bin", "model.added_attn.25.output_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.26.attn.k_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.26.attn.o_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.26.attn.q_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.26.attn.v_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.26.input_proj.bias": "pytorch_model-00006-of-00006.bin", "model.added_attn.26.input_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.26.output_proj.bias": "pytorch_model-00006-of-00006.bin", "model.added_attn.26.output_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.27.attn.k_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.27.attn.o_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.27.attn.q_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.27.attn.v_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.27.input_proj.bias": "pytorch_model-00006-of-00006.bin", "model.added_attn.27.input_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.27.output_proj.bias": "pytorch_model-00006-of-00006.bin", "model.added_attn.27.output_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.28.attn.k_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.28.attn.o_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.28.attn.q_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.28.attn.v_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.28.input_proj.bias": "pytorch_model-00006-of-00006.bin", "model.added_attn.28.input_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.28.output_proj.bias": "pytorch_model-00006-of-00006.bin", "model.added_attn.28.output_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.29.attn.k_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.29.attn.o_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.29.attn.q_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.29.attn.v_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.29.input_proj.bias": "pytorch_model-00006-of-00006.bin", "model.added_attn.29.input_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.29.output_proj.bias": "pytorch_model-00006-of-00006.bin", "model.added_attn.29.output_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.3.attn.k_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.3.attn.o_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.3.attn.q_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.3.attn.v_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.3.input_proj.bias": "pytorch_model-00006-of-00006.bin", "model.added_attn.3.input_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.3.output_proj.bias": "pytorch_model-00006-of-00006.bin", "model.added_attn.3.output_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.30.attn.k_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.30.attn.o_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.30.attn.q_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.30.attn.v_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.30.input_proj.bias": "pytorch_model-00006-of-00006.bin", "model.added_attn.30.input_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.30.output_proj.bias": "pytorch_model-00006-of-00006.bin", "model.added_attn.30.output_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.31.attn.k_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.31.attn.o_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.31.attn.q_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.31.attn.v_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.31.input_proj.bias": "pytorch_model-00006-of-00006.bin", "model.added_attn.31.input_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.31.output_proj.bias": "pytorch_model-00006-of-00006.bin", "model.added_attn.31.output_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.32.attn.k_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.32.attn.o_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.32.attn.q_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.32.attn.v_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.32.input_proj.bias": "pytorch_model-00006-of-00006.bin", "model.added_attn.32.input_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.32.output_proj.bias": "pytorch_model-00006-of-00006.bin", "model.added_attn.32.output_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.33.attn.k_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.33.attn.o_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.33.attn.q_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.33.attn.v_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.33.input_proj.bias": "pytorch_model-00006-of-00006.bin", "model.added_attn.33.input_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.33.output_proj.bias": "pytorch_model-00006-of-00006.bin", "model.added_attn.33.output_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.34.attn.k_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.34.attn.o_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.34.attn.q_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.34.attn.v_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.34.input_proj.bias": "pytorch_model-00006-of-00006.bin", "model.added_attn.34.input_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.34.output_proj.bias": "pytorch_model-00006-of-00006.bin", "model.added_attn.34.output_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.35.attn.k_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.35.attn.o_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.35.attn.q_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.35.attn.v_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.35.input_proj.bias": "pytorch_model-00006-of-00006.bin", "model.added_attn.35.input_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.35.output_proj.bias": "pytorch_model-00006-of-00006.bin", "model.added_attn.35.output_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.36.attn.k_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.36.attn.o_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.36.attn.q_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.36.attn.v_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.36.input_proj.bias": "pytorch_model-00006-of-00006.bin", "model.added_attn.36.input_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.36.output_proj.bias": "pytorch_model-00006-of-00006.bin", "model.added_attn.36.output_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.37.attn.k_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.37.attn.o_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.37.attn.q_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.37.attn.v_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.37.input_proj.bias": "pytorch_model-00006-of-00006.bin", "model.added_attn.37.input_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.37.output_proj.bias": "pytorch_model-00006-of-00006.bin", "model.added_attn.37.output_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.38.attn.k_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.38.attn.o_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.38.attn.q_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.38.attn.v_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.38.input_proj.bias": "pytorch_model-00006-of-00006.bin", "model.added_attn.38.input_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.38.output_proj.bias": "pytorch_model-00006-of-00006.bin", "model.added_attn.38.output_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.39.attn.k_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.39.attn.o_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.39.attn.q_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.39.attn.v_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.39.input_proj.bias": "pytorch_model-00006-of-00006.bin", "model.added_attn.39.input_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.39.output_proj.bias": "pytorch_model-00006-of-00006.bin", "model.added_attn.39.output_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.4.attn.k_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.4.attn.o_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.4.attn.q_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.4.attn.v_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.4.input_proj.bias": "pytorch_model-00006-of-00006.bin", "model.added_attn.4.input_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.4.output_proj.bias": "pytorch_model-00006-of-00006.bin", "model.added_attn.4.output_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.5.attn.k_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.5.attn.o_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.5.attn.q_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.5.attn.v_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.5.input_proj.bias": "pytorch_model-00006-of-00006.bin", "model.added_attn.5.input_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.5.output_proj.bias": "pytorch_model-00006-of-00006.bin", "model.added_attn.5.output_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.6.attn.k_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.6.attn.o_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.6.attn.q_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.6.attn.v_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.6.input_proj.bias": "pytorch_model-00006-of-00006.bin", "model.added_attn.6.input_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.6.output_proj.bias": "pytorch_model-00006-of-00006.bin", "model.added_attn.6.output_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.7.attn.k_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.7.attn.o_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.7.attn.q_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.7.attn.v_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.7.input_proj.bias": "pytorch_model-00006-of-00006.bin", "model.added_attn.7.input_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.7.output_proj.bias": "pytorch_model-00006-of-00006.bin", "model.added_attn.7.output_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.8.attn.k_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.8.attn.o_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.8.attn.q_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.8.attn.v_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.8.input_proj.bias": "pytorch_model-00006-of-00006.bin", "model.added_attn.8.input_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.8.output_proj.bias": "pytorch_model-00006-of-00006.bin", "model.added_attn.8.output_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.9.attn.k_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.9.attn.o_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.9.attn.q_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.9.attn.v_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.9.input_proj.bias": "pytorch_model-00006-of-00006.bin", "model.added_attn.9.input_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_attn.9.output_proj.bias": "pytorch_model-00006-of-00006.bin", "model.added_attn.9.output_proj.weight": "pytorch_model-00006-of-00006.bin", "model.added_lns.0.weight": "pytorch_model-00006-of-00006.bin", "model.added_lns.1.weight": "pytorch_model-00006-of-00006.bin", "model.added_lns.10.weight": "pytorch_model-00006-of-00006.bin", "model.added_lns.11.weight": "pytorch_model-00006-of-00006.bin", "model.added_lns.12.weight": "pytorch_model-00006-of-00006.bin", "model.added_lns.13.weight": "pytorch_model-00006-of-00006.bin", "model.added_lns.14.weight": "pytorch_model-00006-of-00006.bin", "model.added_lns.15.weight": "pytorch_model-00006-of-00006.bin", "model.added_lns.16.weight": "pytorch_model-00006-of-00006.bin", "model.added_lns.17.weight": "pytorch_model-00006-of-00006.bin", "model.added_lns.18.weight": "pytorch_model-00006-of-00006.bin", "model.added_lns.19.weight": "pytorch_model-00006-of-00006.bin", "model.added_lns.2.weight": "pytorch_model-00006-of-00006.bin", "model.added_lns.20.weight": "pytorch_model-00006-of-00006.bin", "model.added_lns.21.weight": "pytorch_model-00006-of-00006.bin", "model.added_lns.22.weight": "pytorch_model-00006-of-00006.bin", "model.added_lns.23.weight": "pytorch_model-00006-of-00006.bin", "model.added_lns.24.weight": "pytorch_model-00006-of-00006.bin", "model.added_lns.25.weight": "pytorch_model-00006-of-00006.bin", "model.added_lns.26.weight": "pytorch_model-00006-of-00006.bin", "model.added_lns.27.weight": "pytorch_model-00006-of-00006.bin", "model.added_lns.28.weight": "pytorch_model-00006-of-00006.bin", "model.added_lns.29.weight": "pytorch_model-00006-of-00006.bin", "model.added_lns.3.weight": "pytorch_model-00006-of-00006.bin", "model.added_lns.30.weight": "pytorch_model-00006-of-00006.bin", "model.added_lns.31.weight": "pytorch_model-00006-of-00006.bin", "model.added_lns.32.weight": "pytorch_model-00006-of-00006.bin", "model.added_lns.33.weight": "pytorch_model-00006-of-00006.bin", "model.added_lns.34.weight": "pytorch_model-00006-of-00006.bin", "model.added_lns.35.weight": "pytorch_model-00006-of-00006.bin", "model.added_lns.36.weight": "pytorch_model-00006-of-00006.bin", "model.added_lns.37.weight": "pytorch_model-00006-of-00006.bin", "model.added_lns.38.weight": "pytorch_model-00006-of-00006.bin", "model.added_lns.39.weight": "pytorch_model-00006-of-00006.bin", "model.added_lns.4.weight": "pytorch_model-00006-of-00006.bin", "model.added_lns.5.weight": "pytorch_model-00006-of-00006.bin", "model.added_lns.6.weight": "pytorch_model-00006-of-00006.bin", "model.added_lns.7.weight": "pytorch_model-00006-of-00006.bin", "model.added_lns.8.weight": "pytorch_model-00006-of-00006.bin", "model.added_lns.9.weight": "pytorch_model-00006-of-00006.bin", "model.added_lrmlp.0.mm_a.weight": "pytorch_model-00006-of-00006.bin", "model.added_lrmlp.0.mm_b.weight": "pytorch_model-00006-of-00006.bin", "model.added_lrmlp.1.mm_a.weight": "pytorch_model-00006-of-00006.bin", "model.added_lrmlp.1.mm_b.weight": "pytorch_model-00006-of-00006.bin", "model.added_lrmlp.10.mm_a.weight": "pytorch_model-00006-of-00006.bin", "model.added_lrmlp.10.mm_b.weight": "pytorch_model-00006-of-00006.bin", "model.added_lrmlp.11.mm_a.weight": "pytorch_model-00006-of-00006.bin", "model.added_lrmlp.11.mm_b.weight": "pytorch_model-00006-of-00006.bin", "model.added_lrmlp.12.mm_a.weight": "pytorch_model-00006-of-00006.bin", "model.added_lrmlp.12.mm_b.weight": "pytorch_model-00006-of-00006.bin", "model.added_lrmlp.13.mm_a.weight": "pytorch_model-00006-of-00006.bin", "model.added_lrmlp.13.mm_b.weight": "pytorch_model-00006-of-00006.bin", "model.added_lrmlp.14.mm_a.weight": "pytorch_model-00006-of-00006.bin", "model.added_lrmlp.14.mm_b.weight": "pytorch_model-00006-of-00006.bin", "model.added_lrmlp.15.mm_a.weight": "pytorch_model-00006-of-00006.bin", "model.added_lrmlp.15.mm_b.weight": "pytorch_model-00006-of-00006.bin", "model.added_lrmlp.16.mm_a.weight": "pytorch_model-00006-of-00006.bin", "model.added_lrmlp.16.mm_b.weight": "pytorch_model-00006-of-00006.bin", "model.added_lrmlp.17.mm_a.weight": "pytorch_model-00006-of-00006.bin", "model.added_lrmlp.17.mm_b.weight": "pytorch_model-00006-of-00006.bin", "model.added_lrmlp.18.mm_a.weight": "pytorch_model-00006-of-00006.bin", "model.added_lrmlp.18.mm_b.weight": "pytorch_model-00006-of-00006.bin", "model.added_lrmlp.19.mm_a.weight": "pytorch_model-00006-of-00006.bin", "model.added_lrmlp.19.mm_b.weight": "pytorch_model-00006-of-00006.bin", "model.added_lrmlp.2.mm_a.weight": "pytorch_model-00006-of-00006.bin", "model.added_lrmlp.2.mm_b.weight": "pytorch_model-00006-of-00006.bin", "model.added_lrmlp.20.mm_a.weight": "pytorch_model-00006-of-00006.bin", "model.added_lrmlp.20.mm_b.weight": "pytorch_model-00006-of-00006.bin", "model.added_lrmlp.21.mm_a.weight": "pytorch_model-00006-of-00006.bin", "model.added_lrmlp.21.mm_b.weight": "pytorch_model-00006-of-00006.bin", "model.added_lrmlp.22.mm_a.weight": "pytorch_model-00006-of-00006.bin", "model.added_lrmlp.22.mm_b.weight": "pytorch_model-00006-of-00006.bin", "model.added_lrmlp.23.mm_a.weight": "pytorch_model-00006-of-00006.bin", "model.added_lrmlp.23.mm_b.weight": "pytorch_model-00006-of-00006.bin", "model.added_lrmlp.24.mm_a.weight": "pytorch_model-00006-of-00006.bin", "model.added_lrmlp.24.mm_b.weight": "pytorch_model-00006-of-00006.bin", "model.added_lrmlp.25.mm_a.weight": "pytorch_model-00006-of-00006.bin", "model.added_lrmlp.25.mm_b.weight": "pytorch_model-00006-of-00006.bin", "model.added_lrmlp.26.mm_a.weight": "pytorch_model-00006-of-00006.bin", "model.added_lrmlp.26.mm_b.weight": "pytorch_model-00006-of-00006.bin", "model.added_lrmlp.27.mm_a.weight": "pytorch_model-00006-of-00006.bin", "model.added_lrmlp.27.mm_b.weight": "pytorch_model-00006-of-00006.bin", "model.added_lrmlp.28.mm_a.weight": "pytorch_model-00006-of-00006.bin", "model.added_lrmlp.28.mm_b.weight": "pytorch_model-00006-of-00006.bin", "model.added_lrmlp.29.mm_a.weight": "pytorch_model-00006-of-00006.bin", "model.added_lrmlp.29.mm_b.weight": "pytorch_model-00006-of-00006.bin", "model.added_lrmlp.3.mm_a.weight": "pytorch_model-00006-of-00006.bin", "model.added_lrmlp.3.mm_b.weight": "pytorch_model-00006-of-00006.bin", "model.added_lrmlp.30.mm_a.weight": "pytorch_model-00006-of-00006.bin", "model.added_lrmlp.30.mm_b.weight": "pytorch_model-00006-of-00006.bin", "model.added_lrmlp.31.mm_a.weight": "pytorch_model-00006-of-00006.bin", "model.added_lrmlp.31.mm_b.weight": "pytorch_model-00006-of-00006.bin", "model.added_lrmlp.32.mm_a.weight": "pytorch_model-00006-of-00006.bin", "model.added_lrmlp.32.mm_b.weight": "pytorch_model-00006-of-00006.bin", "model.added_lrmlp.33.mm_a.weight": "pytorch_model-00006-of-00006.bin", "model.added_lrmlp.33.mm_b.weight": "pytorch_model-00006-of-00006.bin", "model.added_lrmlp.34.mm_a.weight": "pytorch_model-00006-of-00006.bin", "model.added_lrmlp.34.mm_b.weight": "pytorch_model-00006-of-00006.bin", "model.added_lrmlp.35.mm_a.weight": "pytorch_model-00006-of-00006.bin", "model.added_lrmlp.35.mm_b.weight": "pytorch_model-00006-of-00006.bin", "model.added_lrmlp.36.mm_a.weight": "pytorch_model-00006-of-00006.bin", "model.added_lrmlp.36.mm_b.weight": "pytorch_model-00006-of-00006.bin", "model.added_lrmlp.37.mm_a.weight": "pytorch_model-00006-of-00006.bin", "model.added_lrmlp.37.mm_b.weight": "pytorch_model-00006-of-00006.bin", "model.added_lrmlp.38.mm_a.weight": "pytorch_model-00006-of-00006.bin", "model.added_lrmlp.38.mm_b.weight": "pytorch_model-00006-of-00006.bin", "model.added_lrmlp.39.mm_a.weight": "pytorch_model-00006-of-00006.bin", "model.added_lrmlp.39.mm_b.weight": "pytorch_model-00006-of-00006.bin", "model.added_lrmlp.4.mm_a.weight": "pytorch_model-00006-of-00006.bin", "model.added_lrmlp.4.mm_b.weight": "pytorch_model-00006-of-00006.bin", "model.added_lrmlp.5.mm_a.weight": "pytorch_model-00006-of-00006.bin", "model.added_lrmlp.5.mm_b.weight": "pytorch_model-00006-of-00006.bin", "model.added_lrmlp.6.mm_a.weight": "pytorch_model-00006-of-00006.bin", "model.added_lrmlp.6.mm_b.weight": "pytorch_model-00006-of-00006.bin", "model.added_lrmlp.7.mm_a.weight": "pytorch_model-00006-of-00006.bin", "model.added_lrmlp.7.mm_b.weight": "pytorch_model-00006-of-00006.bin", "model.added_lrmlp.8.mm_a.weight": "pytorch_model-00006-of-00006.bin", "model.added_lrmlp.8.mm_b.weight": "pytorch_model-00006-of-00006.bin", "model.added_lrmlp.9.mm_a.weight": "pytorch_model-00006-of-00006.bin", "model.added_lrmlp.9.mm_b.weight": "pytorch_model-00006-of-00006.bin", "model.added_output_lns.0.weight": "pytorch_model-00006-of-00006.bin", "model.added_output_lns.1.weight": "pytorch_model-00006-of-00006.bin", "model.added_output_lns.10.weight": "pytorch_model-00006-of-00006.bin", "model.added_output_lns.11.weight": "pytorch_model-00006-of-00006.bin", "model.added_output_lns.12.weight": "pytorch_model-00006-of-00006.bin", "model.added_output_lns.13.weight": "pytorch_model-00006-of-00006.bin", "model.added_output_lns.14.weight": "pytorch_model-00006-of-00006.bin", "model.added_output_lns.15.weight": "pytorch_model-00006-of-00006.bin", "model.added_output_lns.16.weight": "pytorch_model-00006-of-00006.bin", "model.added_output_lns.17.weight": "pytorch_model-00006-of-00006.bin", "model.added_output_lns.18.weight": "pytorch_model-00006-of-00006.bin", "model.added_output_lns.19.weight": "pytorch_model-00006-of-00006.bin", "model.added_output_lns.2.weight": "pytorch_model-00006-of-00006.bin", "model.added_output_lns.20.weight": "pytorch_model-00006-of-00006.bin", "model.added_output_lns.21.weight": "pytorch_model-00006-of-00006.bin", "model.added_output_lns.22.weight": "pytorch_model-00006-of-00006.bin", "model.added_output_lns.23.weight": "pytorch_model-00006-of-00006.bin", "model.added_output_lns.24.weight": "pytorch_model-00006-of-00006.bin", "model.added_output_lns.25.weight": "pytorch_model-00006-of-00006.bin", "model.added_output_lns.26.weight": "pytorch_model-00006-of-00006.bin", "model.added_output_lns.27.weight": "pytorch_model-00006-of-00006.bin", "model.added_output_lns.28.weight": "pytorch_model-00006-of-00006.bin", "model.added_output_lns.29.weight": "pytorch_model-00006-of-00006.bin", "model.added_output_lns.3.weight": "pytorch_model-00006-of-00006.bin", "model.added_output_lns.30.weight": "pytorch_model-00006-of-00006.bin", "model.added_output_lns.31.weight": "pytorch_model-00006-of-00006.bin", "model.added_output_lns.32.weight": "pytorch_model-00006-of-00006.bin", "model.added_output_lns.33.weight": "pytorch_model-00006-of-00006.bin", "model.added_output_lns.34.weight": "pytorch_model-00006-of-00006.bin", "model.added_output_lns.35.weight": "pytorch_model-00006-of-00006.bin", "model.added_output_lns.36.weight": "pytorch_model-00006-of-00006.bin", "model.added_output_lns.37.weight": "pytorch_model-00006-of-00006.bin", "model.added_output_lns.38.weight": "pytorch_model-00006-of-00006.bin", "model.added_output_lns.39.weight": "pytorch_model-00006-of-00006.bin", "model.added_output_lns.4.weight": "pytorch_model-00006-of-00006.bin", "model.added_output_lns.5.weight": "pytorch_model-00006-of-00006.bin", "model.added_output_lns.6.weight": "pytorch_model-00006-of-00006.bin", "model.added_output_lns.7.weight": "pytorch_model-00006-of-00006.bin", "model.added_output_lns.8.weight": "pytorch_model-00006-of-00006.bin", "model.added_output_lns.9.weight": "pytorch_model-00006-of-00006.bin", "model.embed_tokens.weight": "pytorch_model-00001-of-00006.bin", "model.layers.0.input_layernorm.weight": "pytorch_model-00001-of-00006.bin", "model.layers.0.mlp.down_proj.weight": "pytorch_model-00001-of-00006.bin", "model.layers.0.mlp.gate_proj.weight": "pytorch_model-00001-of-00006.bin", "model.layers.0.mlp.up_proj.weight": "pytorch_model-00001-of-00006.bin", "model.layers.0.post_attention_layernorm.weight": "pytorch_model-00001-of-00006.bin", "model.layers.0.self_attn.k_proj.weight": "pytorch_model-00001-of-00006.bin", "model.layers.0.self_attn.o_proj.weight": "pytorch_model-00001-of-00006.bin", "model.layers.0.self_attn.q_proj.weight": "pytorch_model-00001-of-00006.bin", "model.layers.0.self_attn.v_proj.weight": "pytorch_model-00001-of-00006.bin", "model.layers.1.input_layernorm.weight": "pytorch_model-00001-of-00006.bin", "model.layers.1.mlp.down_proj.weight": "pytorch_model-00001-of-00006.bin", "model.layers.1.mlp.gate_proj.weight": "pytorch_model-00001-of-00006.bin", "model.layers.1.mlp.up_proj.weight": "pytorch_model-00001-of-00006.bin", "model.layers.1.post_attention_layernorm.weight": "pytorch_model-00001-of-00006.bin", "model.layers.1.self_attn.k_proj.weight": "pytorch_model-00001-of-00006.bin", "model.layers.1.self_attn.o_proj.weight": "pytorch_model-00001-of-00006.bin", "model.layers.1.self_attn.q_proj.weight": "pytorch_model-00001-of-00006.bin", "model.layers.1.self_attn.v_proj.weight": "pytorch_model-00001-of-00006.bin", "model.layers.10.input_layernorm.weight": "pytorch_model-00002-of-00006.bin", "model.layers.10.mlp.down_proj.weight": "pytorch_model-00002-of-00006.bin", "model.layers.10.mlp.gate_proj.weight": "pytorch_model-00002-of-00006.bin", "model.layers.10.mlp.up_proj.weight": "pytorch_model-00002-of-00006.bin", "model.layers.10.post_attention_layernorm.weight": "pytorch_model-00002-of-00006.bin", "model.layers.10.self_attn.k_proj.weight": "pytorch_model-00002-of-00006.bin", "model.layers.10.self_attn.o_proj.weight": "pytorch_model-00002-of-00006.bin", "model.layers.10.self_attn.q_proj.weight": "pytorch_model-00002-of-00006.bin", "model.layers.10.self_attn.v_proj.weight": "pytorch_model-00002-of-00006.bin", "model.layers.11.input_layernorm.weight": "pytorch_model-00002-of-00006.bin", "model.layers.11.mlp.down_proj.weight": "pytorch_model-00002-of-00006.bin", "model.layers.11.mlp.gate_proj.weight": "pytorch_model-00002-of-00006.bin", "model.layers.11.mlp.up_proj.weight": "pytorch_model-00002-of-00006.bin", "model.layers.11.post_attention_layernorm.weight": "pytorch_model-00002-of-00006.bin", "model.layers.11.self_attn.k_proj.weight": "pytorch_model-00002-of-00006.bin", "model.layers.11.self_attn.o_proj.weight": "pytorch_model-00002-of-00006.bin", "model.layers.11.self_attn.q_proj.weight": "pytorch_model-00002-of-00006.bin", "model.layers.11.self_attn.v_proj.weight": "pytorch_model-00002-of-00006.bin", "model.layers.12.input_layernorm.weight": "pytorch_model-00002-of-00006.bin", "model.layers.12.mlp.down_proj.weight": "pytorch_model-00002-of-00006.bin", "model.layers.12.mlp.gate_proj.weight": "pytorch_model-00002-of-00006.bin", "model.layers.12.mlp.up_proj.weight": "pytorch_model-00002-of-00006.bin", "model.layers.12.post_attention_layernorm.weight": "pytorch_model-00002-of-00006.bin", "model.layers.12.self_attn.k_proj.weight": "pytorch_model-00002-of-00006.bin", "model.layers.12.self_attn.o_proj.weight": "pytorch_model-00002-of-00006.bin", "model.layers.12.self_attn.q_proj.weight": "pytorch_model-00002-of-00006.bin", "model.layers.12.self_attn.v_proj.weight": "pytorch_model-00002-of-00006.bin", "model.layers.13.input_layernorm.weight": "pytorch_model-00002-of-00006.bin", "model.layers.13.mlp.down_proj.weight": "pytorch_model-00002-of-00006.bin", "model.layers.13.mlp.gate_proj.weight": "pytorch_model-00002-of-00006.bin", "model.layers.13.mlp.up_proj.weight": "pytorch_model-00002-of-00006.bin", "model.layers.13.post_attention_layernorm.weight": "pytorch_model-00002-of-00006.bin", "model.layers.13.self_attn.k_proj.weight": "pytorch_model-00002-of-00006.bin", "model.layers.13.self_attn.o_proj.weight": "pytorch_model-00002-of-00006.bin", "model.layers.13.self_attn.q_proj.weight": "pytorch_model-00002-of-00006.bin", "model.layers.13.self_attn.v_proj.weight": "pytorch_model-00002-of-00006.bin", "model.layers.14.input_layernorm.weight": "pytorch_model-00002-of-00006.bin", "model.layers.14.mlp.down_proj.weight": "pytorch_model-00002-of-00006.bin", "model.layers.14.mlp.gate_proj.weight": "pytorch_model-00002-of-00006.bin", "model.layers.14.mlp.up_proj.weight": "pytorch_model-00002-of-00006.bin", "model.layers.14.post_attention_layernorm.weight": "pytorch_model-00002-of-00006.bin", "model.layers.14.self_attn.k_proj.weight": "pytorch_model-00002-of-00006.bin", "model.layers.14.self_attn.o_proj.weight": "pytorch_model-00002-of-00006.bin", "model.layers.14.self_attn.q_proj.weight": "pytorch_model-00002-of-00006.bin", "model.layers.14.self_attn.v_proj.weight": "pytorch_model-00002-of-00006.bin", "model.layers.15.input_layernorm.weight": "pytorch_model-00003-of-00006.bin", "model.layers.15.mlp.down_proj.weight": "pytorch_model-00003-of-00006.bin", "model.layers.15.mlp.gate_proj.weight": "pytorch_model-00003-of-00006.bin", "model.layers.15.mlp.up_proj.weight": "pytorch_model-00003-of-00006.bin", "model.layers.15.post_attention_layernorm.weight": "pytorch_model-00003-of-00006.bin", "model.layers.15.self_attn.k_proj.weight": "pytorch_model-00002-of-00006.bin", "model.layers.15.self_attn.o_proj.weight": "pytorch_model-00003-of-00006.bin", "model.layers.15.self_attn.q_proj.weight": "pytorch_model-00002-of-00006.bin", "model.layers.15.self_attn.v_proj.weight": "pytorch_model-00003-of-00006.bin", "model.layers.16.input_layernorm.weight": "pytorch_model-00003-of-00006.bin", "model.layers.16.mlp.down_proj.weight": "pytorch_model-00003-of-00006.bin", "model.layers.16.mlp.gate_proj.weight": "pytorch_model-00003-of-00006.bin", "model.layers.16.mlp.up_proj.weight": "pytorch_model-00003-of-00006.bin", "model.layers.16.post_attention_layernorm.weight": "pytorch_model-00003-of-00006.bin", "model.layers.16.self_attn.k_proj.weight": "pytorch_model-00003-of-00006.bin", "model.layers.16.self_attn.o_proj.weight": "pytorch_model-00003-of-00006.bin", "model.layers.16.self_attn.q_proj.weight": "pytorch_model-00003-of-00006.bin", "model.layers.16.self_attn.v_proj.weight": "pytorch_model-00003-of-00006.bin", "model.layers.17.input_layernorm.weight": "pytorch_model-00003-of-00006.bin", "model.layers.17.mlp.down_proj.weight": "pytorch_model-00003-of-00006.bin", "model.layers.17.mlp.gate_proj.weight": "pytorch_model-00003-of-00006.bin", "model.layers.17.mlp.up_proj.weight": "pytorch_model-00003-of-00006.bin", "model.layers.17.post_attention_layernorm.weight": "pytorch_model-00003-of-00006.bin", "model.layers.17.self_attn.k_proj.weight": "pytorch_model-00003-of-00006.bin", "model.layers.17.self_attn.o_proj.weight": "pytorch_model-00003-of-00006.bin", "model.layers.17.self_attn.q_proj.weight": "pytorch_model-00003-of-00006.bin", "model.layers.17.self_attn.v_proj.weight": "pytorch_model-00003-of-00006.bin", "model.layers.18.input_layernorm.weight": "pytorch_model-00003-of-00006.bin", "model.layers.18.mlp.down_proj.weight": "pytorch_model-00003-of-00006.bin", "model.layers.18.mlp.gate_proj.weight": "pytorch_model-00003-of-00006.bin", "model.layers.18.mlp.up_proj.weight": "pytorch_model-00003-of-00006.bin", "model.layers.18.post_attention_layernorm.weight": "pytorch_model-00003-of-00006.bin", "model.layers.18.self_attn.k_proj.weight": "pytorch_model-00003-of-00006.bin", "model.layers.18.self_attn.o_proj.weight": "pytorch_model-00003-of-00006.bin", "model.layers.18.self_attn.q_proj.weight": "pytorch_model-00003-of-00006.bin", "model.layers.18.self_attn.v_proj.weight": "pytorch_model-00003-of-00006.bin", "model.layers.19.input_layernorm.weight": "pytorch_model-00003-of-00006.bin", "model.layers.19.mlp.down_proj.weight": "pytorch_model-00003-of-00006.bin", "model.layers.19.mlp.gate_proj.weight": "pytorch_model-00003-of-00006.bin", "model.layers.19.mlp.up_proj.weight": "pytorch_model-00003-of-00006.bin", "model.layers.19.post_attention_layernorm.weight": "pytorch_model-00003-of-00006.bin", "model.layers.19.self_attn.k_proj.weight": "pytorch_model-00003-of-00006.bin", "model.layers.19.self_attn.o_proj.weight": "pytorch_model-00003-of-00006.bin", "model.layers.19.self_attn.q_proj.weight": "pytorch_model-00003-of-00006.bin", "model.layers.19.self_attn.v_proj.weight": "pytorch_model-00003-of-00006.bin", "model.layers.2.input_layernorm.weight": "pytorch_model-00001-of-00006.bin", "model.layers.2.mlp.down_proj.weight": "pytorch_model-00001-of-00006.bin", "model.layers.2.mlp.gate_proj.weight": "pytorch_model-00001-of-00006.bin", "model.layers.2.mlp.up_proj.weight": "pytorch_model-00001-of-00006.bin", "model.layers.2.post_attention_layernorm.weight": "pytorch_model-00001-of-00006.bin", "model.layers.2.self_attn.k_proj.weight": "pytorch_model-00001-of-00006.bin", "model.layers.2.self_attn.o_proj.weight": "pytorch_model-00001-of-00006.bin", "model.layers.2.self_attn.q_proj.weight": "pytorch_model-00001-of-00006.bin", "model.layers.2.self_attn.v_proj.weight": "pytorch_model-00001-of-00006.bin", "model.layers.20.input_layernorm.weight": "pytorch_model-00003-of-00006.bin", "model.layers.20.mlp.down_proj.weight": "pytorch_model-00003-of-00006.bin", "model.layers.20.mlp.gate_proj.weight": "pytorch_model-00003-of-00006.bin", "model.layers.20.mlp.up_proj.weight": "pytorch_model-00003-of-00006.bin", "model.layers.20.post_attention_layernorm.weight": "pytorch_model-00003-of-00006.bin", "model.layers.20.self_attn.k_proj.weight": "pytorch_model-00003-of-00006.bin", "model.layers.20.self_attn.o_proj.weight": "pytorch_model-00003-of-00006.bin", "model.layers.20.self_attn.q_proj.weight": "pytorch_model-00003-of-00006.bin", "model.layers.20.self_attn.v_proj.weight": "pytorch_model-00003-of-00006.bin", "model.layers.21.input_layernorm.weight": "pytorch_model-00003-of-00006.bin", "model.layers.21.mlp.down_proj.weight": "pytorch_model-00003-of-00006.bin", "model.layers.21.mlp.gate_proj.weight": "pytorch_model-00003-of-00006.bin", "model.layers.21.mlp.up_proj.weight": "pytorch_model-00003-of-00006.bin", "model.layers.21.post_attention_layernorm.weight": "pytorch_model-00003-of-00006.bin", "model.layers.21.self_attn.k_proj.weight": "pytorch_model-00003-of-00006.bin", "model.layers.21.self_attn.o_proj.weight": "pytorch_model-00003-of-00006.bin", "model.layers.21.self_attn.q_proj.weight": "pytorch_model-00003-of-00006.bin", "model.layers.21.self_attn.v_proj.weight": "pytorch_model-00003-of-00006.bin", "model.layers.22.input_layernorm.weight": "pytorch_model-00003-of-00006.bin", "model.layers.22.mlp.down_proj.weight": "pytorch_model-00003-of-00006.bin", "model.layers.22.mlp.gate_proj.weight": "pytorch_model-00003-of-00006.bin", "model.layers.22.mlp.up_proj.weight": "pytorch_model-00003-of-00006.bin", "model.layers.22.post_attention_layernorm.weight": "pytorch_model-00003-of-00006.bin", "model.layers.22.self_attn.k_proj.weight": "pytorch_model-00003-of-00006.bin", "model.layers.22.self_attn.o_proj.weight": "pytorch_model-00003-of-00006.bin", "model.layers.22.self_attn.q_proj.weight": "pytorch_model-00003-of-00006.bin", "model.layers.22.self_attn.v_proj.weight": "pytorch_model-00003-of-00006.bin", "model.layers.23.input_layernorm.weight": "pytorch_model-00004-of-00006.bin", "model.layers.23.mlp.down_proj.weight": "pytorch_model-00004-of-00006.bin", "model.layers.23.mlp.gate_proj.weight": "pytorch_model-00004-of-00006.bin", "model.layers.23.mlp.up_proj.weight": "pytorch_model-00004-of-00006.bin", "model.layers.23.post_attention_layernorm.weight": "pytorch_model-00004-of-00006.bin", "model.layers.23.self_attn.k_proj.weight": "pytorch_model-00004-of-00006.bin", "model.layers.23.self_attn.o_proj.weight": "pytorch_model-00004-of-00006.bin", "model.layers.23.self_attn.q_proj.weight": "pytorch_model-00004-of-00006.bin", "model.layers.23.self_attn.v_proj.weight": "pytorch_model-00004-of-00006.bin", "model.layers.24.input_layernorm.weight": "pytorch_model-00004-of-00006.bin", "model.layers.24.mlp.down_proj.weight": "pytorch_model-00004-of-00006.bin", "model.layers.24.mlp.gate_proj.weight": "pytorch_model-00004-of-00006.bin", "model.layers.24.mlp.up_proj.weight": "pytorch_model-00004-of-00006.bin", "model.layers.24.post_attention_layernorm.weight": "pytorch_model-00004-of-00006.bin", "model.layers.24.self_attn.k_proj.weight": "pytorch_model-00004-of-00006.bin", "model.layers.24.self_attn.o_proj.weight": "pytorch_model-00004-of-00006.bin", "model.layers.24.self_attn.q_proj.weight": "pytorch_model-00004-of-00006.bin", "model.layers.24.self_attn.v_proj.weight": "pytorch_model-00004-of-00006.bin", "model.layers.25.input_layernorm.weight": "pytorch_model-00004-of-00006.bin", "model.layers.25.mlp.down_proj.weight": "pytorch_model-00004-of-00006.bin", "model.layers.25.mlp.gate_proj.weight": "pytorch_model-00004-of-00006.bin", "model.layers.25.mlp.up_proj.weight": "pytorch_model-00004-of-00006.bin", "model.layers.25.post_attention_layernorm.weight": "pytorch_model-00004-of-00006.bin", "model.layers.25.self_attn.k_proj.weight": "pytorch_model-00004-of-00006.bin", "model.layers.25.self_attn.o_proj.weight": "pytorch_model-00004-of-00006.bin", "model.layers.25.self_attn.q_proj.weight": "pytorch_model-00004-of-00006.bin", "model.layers.25.self_attn.v_proj.weight": "pytorch_model-00004-of-00006.bin", "model.layers.26.input_layernorm.weight": "pytorch_model-00004-of-00006.bin", "model.layers.26.mlp.down_proj.weight": "pytorch_model-00004-of-00006.bin", "model.layers.26.mlp.gate_proj.weight": "pytorch_model-00004-of-00006.bin", "model.layers.26.mlp.up_proj.weight": "pytorch_model-00004-of-00006.bin", "model.layers.26.post_attention_layernorm.weight": "pytorch_model-00004-of-00006.bin", "model.layers.26.self_attn.k_proj.weight": "pytorch_model-00004-of-00006.bin", "model.layers.26.self_attn.o_proj.weight": "pytorch_model-00004-of-00006.bin", "model.layers.26.self_attn.q_proj.weight": "pytorch_model-00004-of-00006.bin", "model.layers.26.self_attn.v_proj.weight": "pytorch_model-00004-of-00006.bin", "model.layers.27.input_layernorm.weight": "pytorch_model-00004-of-00006.bin", "model.layers.27.mlp.down_proj.weight": "pytorch_model-00004-of-00006.bin", "model.layers.27.mlp.gate_proj.weight": "pytorch_model-00004-of-00006.bin", "model.layers.27.mlp.up_proj.weight": "pytorch_model-00004-of-00006.bin", "model.layers.27.post_attention_layernorm.weight": "pytorch_model-00004-of-00006.bin", "model.layers.27.self_attn.k_proj.weight": "pytorch_model-00004-of-00006.bin", "model.layers.27.self_attn.o_proj.weight": "pytorch_model-00004-of-00006.bin", "model.layers.27.self_attn.q_proj.weight": "pytorch_model-00004-of-00006.bin", "model.layers.27.self_attn.v_proj.weight": "pytorch_model-00004-of-00006.bin", "model.layers.28.input_layernorm.weight": "pytorch_model-00004-of-00006.bin", "model.layers.28.mlp.down_proj.weight": "pytorch_model-00004-of-00006.bin", "model.layers.28.mlp.gate_proj.weight": "pytorch_model-00004-of-00006.bin", "model.layers.28.mlp.up_proj.weight": "pytorch_model-00004-of-00006.bin", "model.layers.28.post_attention_layernorm.weight": "pytorch_model-00004-of-00006.bin", "model.layers.28.self_attn.k_proj.weight": "pytorch_model-00004-of-00006.bin", "model.layers.28.self_attn.o_proj.weight": "pytorch_model-00004-of-00006.bin", "model.layers.28.self_attn.q_proj.weight": "pytorch_model-00004-of-00006.bin", "model.layers.28.self_attn.v_proj.weight": "pytorch_model-00004-of-00006.bin", "model.layers.29.input_layernorm.weight": "pytorch_model-00004-of-00006.bin", "model.layers.29.mlp.down_proj.weight": "pytorch_model-00004-of-00006.bin", "model.layers.29.mlp.gate_proj.weight": "pytorch_model-00004-of-00006.bin", "model.layers.29.mlp.up_proj.weight": "pytorch_model-00004-of-00006.bin", "model.layers.29.post_attention_layernorm.weight": "pytorch_model-00004-of-00006.bin", "model.layers.29.self_attn.k_proj.weight": "pytorch_model-00004-of-00006.bin", "model.layers.29.self_attn.o_proj.weight": "pytorch_model-00004-of-00006.bin", "model.layers.29.self_attn.q_proj.weight": "pytorch_model-00004-of-00006.bin", "model.layers.29.self_attn.v_proj.weight": "pytorch_model-00004-of-00006.bin", "model.layers.3.input_layernorm.weight": "pytorch_model-00001-of-00006.bin", "model.layers.3.mlp.down_proj.weight": "pytorch_model-00001-of-00006.bin", "model.layers.3.mlp.gate_proj.weight": "pytorch_model-00001-of-00006.bin", "model.layers.3.mlp.up_proj.weight": "pytorch_model-00001-of-00006.bin", "model.layers.3.post_attention_layernorm.weight": "pytorch_model-00001-of-00006.bin", "model.layers.3.self_attn.k_proj.weight": "pytorch_model-00001-of-00006.bin", "model.layers.3.self_attn.o_proj.weight": "pytorch_model-00001-of-00006.bin", "model.layers.3.self_attn.q_proj.weight": "pytorch_model-00001-of-00006.bin", "model.layers.3.self_attn.v_proj.weight": "pytorch_model-00001-of-00006.bin", "model.layers.30.input_layernorm.weight": "pytorch_model-00005-of-00006.bin", "model.layers.30.mlp.down_proj.weight": "pytorch_model-00005-of-00006.bin", "model.layers.30.mlp.gate_proj.weight": "pytorch_model-00004-of-00006.bin", "model.layers.30.mlp.up_proj.weight": "pytorch_model-00004-of-00006.bin", "model.layers.30.post_attention_layernorm.weight": "pytorch_model-00005-of-00006.bin", "model.layers.30.self_attn.k_proj.weight": "pytorch_model-00004-of-00006.bin", "model.layers.30.self_attn.o_proj.weight": "pytorch_model-00004-of-00006.bin", "model.layers.30.self_attn.q_proj.weight": "pytorch_model-00004-of-00006.bin", "model.layers.30.self_attn.v_proj.weight": "pytorch_model-00004-of-00006.bin", "model.layers.31.input_layernorm.weight": "pytorch_model-00005-of-00006.bin", "model.layers.31.mlp.down_proj.weight": "pytorch_model-00005-of-00006.bin", "model.layers.31.mlp.gate_proj.weight": "pytorch_model-00005-of-00006.bin", "model.layers.31.mlp.up_proj.weight": "pytorch_model-00005-of-00006.bin", "model.layers.31.post_attention_layernorm.weight": "pytorch_model-00005-of-00006.bin", "model.layers.31.self_attn.k_proj.weight": "pytorch_model-00005-of-00006.bin", "model.layers.31.self_attn.o_proj.weight": "pytorch_model-00005-of-00006.bin", "model.layers.31.self_attn.q_proj.weight": "pytorch_model-00005-of-00006.bin", "model.layers.31.self_attn.v_proj.weight": "pytorch_model-00005-of-00006.bin", "model.layers.32.input_layernorm.weight": "pytorch_model-00005-of-00006.bin", "model.layers.32.mlp.down_proj.weight": "pytorch_model-00005-of-00006.bin", "model.layers.32.mlp.gate_proj.weight": "pytorch_model-00005-of-00006.bin", "model.layers.32.mlp.up_proj.weight": "pytorch_model-00005-of-00006.bin", "model.layers.32.post_attention_layernorm.weight": "pytorch_model-00005-of-00006.bin", "model.layers.32.self_attn.k_proj.weight": "pytorch_model-00005-of-00006.bin", "model.layers.32.self_attn.o_proj.weight": "pytorch_model-00005-of-00006.bin", "model.layers.32.self_attn.q_proj.weight": "pytorch_model-00005-of-00006.bin", "model.layers.32.self_attn.v_proj.weight": "pytorch_model-00005-of-00006.bin", "model.layers.33.input_layernorm.weight": "pytorch_model-00005-of-00006.bin", "model.layers.33.mlp.down_proj.weight": "pytorch_model-00005-of-00006.bin", "model.layers.33.mlp.gate_proj.weight": "pytorch_model-00005-of-00006.bin", "model.layers.33.mlp.up_proj.weight": "pytorch_model-00005-of-00006.bin", "model.layers.33.post_attention_layernorm.weight": "pytorch_model-00005-of-00006.bin", "model.layers.33.self_attn.k_proj.weight": "pytorch_model-00005-of-00006.bin", "model.layers.33.self_attn.o_proj.weight": "pytorch_model-00005-of-00006.bin", "model.layers.33.self_attn.q_proj.weight": "pytorch_model-00005-of-00006.bin", "model.layers.33.self_attn.v_proj.weight": "pytorch_model-00005-of-00006.bin", "model.layers.34.input_layernorm.weight": "pytorch_model-00005-of-00006.bin", "model.layers.34.mlp.down_proj.weight": "pytorch_model-00005-of-00006.bin", "model.layers.34.mlp.gate_proj.weight": "pytorch_model-00005-of-00006.bin", "model.layers.34.mlp.up_proj.weight": "pytorch_model-00005-of-00006.bin", "model.layers.34.post_attention_layernorm.weight": "pytorch_model-00005-of-00006.bin", "model.layers.34.self_attn.k_proj.weight": "pytorch_model-00005-of-00006.bin", "model.layers.34.self_attn.o_proj.weight": "pytorch_model-00005-of-00006.bin", "model.layers.34.self_attn.q_proj.weight": "pytorch_model-00005-of-00006.bin", "model.layers.34.self_attn.v_proj.weight": "pytorch_model-00005-of-00006.bin", "model.layers.35.input_layernorm.weight": "pytorch_model-00005-of-00006.bin", "model.layers.35.mlp.down_proj.weight": "pytorch_model-00005-of-00006.bin", "model.layers.35.mlp.gate_proj.weight": "pytorch_model-00005-of-00006.bin", "model.layers.35.mlp.up_proj.weight": "pytorch_model-00005-of-00006.bin", "model.layers.35.post_attention_layernorm.weight": "pytorch_model-00005-of-00006.bin", "model.layers.35.self_attn.k_proj.weight": "pytorch_model-00005-of-00006.bin", "model.layers.35.self_attn.o_proj.weight": "pytorch_model-00005-of-00006.bin", "model.layers.35.self_attn.q_proj.weight": "pytorch_model-00005-of-00006.bin", "model.layers.35.self_attn.v_proj.weight": "pytorch_model-00005-of-00006.bin", "model.layers.36.input_layernorm.weight": "pytorch_model-00005-of-00006.bin", "model.layers.36.mlp.down_proj.weight": "pytorch_model-00005-of-00006.bin", "model.layers.36.mlp.gate_proj.weight": "pytorch_model-00005-of-00006.bin", "model.layers.36.mlp.up_proj.weight": "pytorch_model-00005-of-00006.bin", "model.layers.36.post_attention_layernorm.weight": "pytorch_model-00005-of-00006.bin", "model.layers.36.self_attn.k_proj.weight": "pytorch_model-00005-of-00006.bin", "model.layers.36.self_attn.o_proj.weight": "pytorch_model-00005-of-00006.bin", "model.layers.36.self_attn.q_proj.weight": "pytorch_model-00005-of-00006.bin", "model.layers.36.self_attn.v_proj.weight": "pytorch_model-00005-of-00006.bin", "model.layers.37.input_layernorm.weight": "pytorch_model-00005-of-00006.bin", "model.layers.37.mlp.down_proj.weight": "pytorch_model-00005-of-00006.bin", "model.layers.37.mlp.gate_proj.weight": "pytorch_model-00005-of-00006.bin", "model.layers.37.mlp.up_proj.weight": "pytorch_model-00005-of-00006.bin", "model.layers.37.post_attention_layernorm.weight": "pytorch_model-00005-of-00006.bin", "model.layers.37.self_attn.k_proj.weight": "pytorch_model-00005-of-00006.bin", "model.layers.37.self_attn.o_proj.weight": "pytorch_model-00005-of-00006.bin", "model.layers.37.self_attn.q_proj.weight": "pytorch_model-00005-of-00006.bin", "model.layers.37.self_attn.v_proj.weight": "pytorch_model-00005-of-00006.bin", "model.layers.38.input_layernorm.weight": "pytorch_model-00006-of-00006.bin", "model.layers.38.mlp.down_proj.weight": "pytorch_model-00006-of-00006.bin", "model.layers.38.mlp.gate_proj.weight": "pytorch_model-00005-of-00006.bin", "model.layers.38.mlp.up_proj.weight": "pytorch_model-00006-of-00006.bin", "model.layers.38.post_attention_layernorm.weight": "pytorch_model-00006-of-00006.bin", "model.layers.38.self_attn.k_proj.weight": "pytorch_model-00005-of-00006.bin", "model.layers.38.self_attn.o_proj.weight": "pytorch_model-00005-of-00006.bin", "model.layers.38.self_attn.q_proj.weight": "pytorch_model-00005-of-00006.bin", "model.layers.38.self_attn.v_proj.weight": "pytorch_model-00005-of-00006.bin", "model.layers.39.input_layernorm.weight": "pytorch_model-00006-of-00006.bin", "model.layers.39.mlp.down_proj.weight": "pytorch_model-00006-of-00006.bin", "model.layers.39.mlp.gate_proj.weight": "pytorch_model-00006-of-00006.bin", "model.layers.39.mlp.up_proj.weight": "pytorch_model-00006-of-00006.bin", "model.layers.39.post_attention_layernorm.weight": "pytorch_model-00006-of-00006.bin", "model.layers.39.self_attn.k_proj.weight": "pytorch_model-00006-of-00006.bin", "model.layers.39.self_attn.o_proj.weight": "pytorch_model-00006-of-00006.bin", "model.layers.39.self_attn.q_proj.weight": "pytorch_model-00006-of-00006.bin", "model.layers.39.self_attn.v_proj.weight": "pytorch_model-00006-of-00006.bin", "model.layers.4.input_layernorm.weight": "pytorch_model-00001-of-00006.bin", "model.layers.4.mlp.down_proj.weight": "pytorch_model-00001-of-00006.bin", "model.layers.4.mlp.gate_proj.weight": "pytorch_model-00001-of-00006.bin", "model.layers.4.mlp.up_proj.weight": "pytorch_model-00001-of-00006.bin", "model.layers.4.post_attention_layernorm.weight": "pytorch_model-00001-of-00006.bin", "model.layers.4.self_attn.k_proj.weight": "pytorch_model-00001-of-00006.bin", "model.layers.4.self_attn.o_proj.weight": "pytorch_model-00001-of-00006.bin", "model.layers.4.self_attn.q_proj.weight": "pytorch_model-00001-of-00006.bin", "model.layers.4.self_attn.v_proj.weight": "pytorch_model-00001-of-00006.bin", "model.layers.5.input_layernorm.weight": "pytorch_model-00001-of-00006.bin", "model.layers.5.mlp.down_proj.weight": "pytorch_model-00001-of-00006.bin", "model.layers.5.mlp.gate_proj.weight": "pytorch_model-00001-of-00006.bin", "model.layers.5.mlp.up_proj.weight": "pytorch_model-00001-of-00006.bin", "model.layers.5.post_attention_layernorm.weight": "pytorch_model-00001-of-00006.bin", "model.layers.5.self_attn.k_proj.weight": "pytorch_model-00001-of-00006.bin", "model.layers.5.self_attn.o_proj.weight": "pytorch_model-00001-of-00006.bin", "model.layers.5.self_attn.q_proj.weight": "pytorch_model-00001-of-00006.bin", "model.layers.5.self_attn.v_proj.weight": "pytorch_model-00001-of-00006.bin", "model.layers.6.input_layernorm.weight": "pytorch_model-00001-of-00006.bin", "model.layers.6.mlp.down_proj.weight": "pytorch_model-00001-of-00006.bin", "model.layers.6.mlp.gate_proj.weight": "pytorch_model-00001-of-00006.bin", "model.layers.6.mlp.up_proj.weight": "pytorch_model-00001-of-00006.bin", "model.layers.6.post_attention_layernorm.weight": "pytorch_model-00001-of-00006.bin", "model.layers.6.self_attn.k_proj.weight": "pytorch_model-00001-of-00006.bin", "model.layers.6.self_attn.o_proj.weight": "pytorch_model-00001-of-00006.bin", "model.layers.6.self_attn.q_proj.weight": "pytorch_model-00001-of-00006.bin", "model.layers.6.self_attn.v_proj.weight": "pytorch_model-00001-of-00006.bin", "model.layers.7.input_layernorm.weight": "pytorch_model-00002-of-00006.bin", "model.layers.7.mlp.down_proj.weight": "pytorch_model-00002-of-00006.bin", "model.layers.7.mlp.gate_proj.weight": "pytorch_model-00002-of-00006.bin", "model.layers.7.mlp.up_proj.weight": "pytorch_model-00002-of-00006.bin", "model.layers.7.post_attention_layernorm.weight": "pytorch_model-00002-of-00006.bin", "model.layers.7.self_attn.k_proj.weight": "pytorch_model-00001-of-00006.bin", "model.layers.7.self_attn.o_proj.weight": "pytorch_model-00001-of-00006.bin", "model.layers.7.self_attn.q_proj.weight": "pytorch_model-00001-of-00006.bin", "model.layers.7.self_attn.v_proj.weight": "pytorch_model-00001-of-00006.bin", "model.layers.8.input_layernorm.weight": "pytorch_model-00002-of-00006.bin", "model.layers.8.mlp.down_proj.weight": "pytorch_model-00002-of-00006.bin", "model.layers.8.mlp.gate_proj.weight": "pytorch_model-00002-of-00006.bin", "model.layers.8.mlp.up_proj.weight": "pytorch_model-00002-of-00006.bin", "model.layers.8.post_attention_layernorm.weight": "pytorch_model-00002-of-00006.bin", "model.layers.8.self_attn.k_proj.weight": "pytorch_model-00002-of-00006.bin", "model.layers.8.self_attn.o_proj.weight": "pytorch_model-00002-of-00006.bin", "model.layers.8.self_attn.q_proj.weight": "pytorch_model-00002-of-00006.bin", "model.layers.8.self_attn.v_proj.weight": "pytorch_model-00002-of-00006.bin", "model.layers.9.input_layernorm.weight": "pytorch_model-00002-of-00006.bin", "model.layers.9.mlp.down_proj.weight": "pytorch_model-00002-of-00006.bin", "model.layers.9.mlp.gate_proj.weight": "pytorch_model-00002-of-00006.bin", "model.layers.9.mlp.up_proj.weight": "pytorch_model-00002-of-00006.bin", "model.layers.9.post_attention_layernorm.weight": "pytorch_model-00002-of-00006.bin", "model.layers.9.self_attn.k_proj.weight": "pytorch_model-00002-of-00006.bin", "model.layers.9.self_attn.o_proj.weight": "pytorch_model-00002-of-00006.bin", "model.layers.9.self_attn.q_proj.weight": "pytorch_model-00002-of-00006.bin", "model.layers.9.self_attn.v_proj.weight": "pytorch_model-00002-of-00006.bin", "model.norm.weight": "pytorch_model-00006-of-00006.bin" } }