gpt-j-reward-model / pytorch_model.bin.index.json
zpn's picture
Upload model
779c83e
raw
history blame
55.3 kB
{
"metadata": {
"total_size": 47610475616.0
},
"weight_map": {
"base_model.lm_head.bias": "pytorch_model-00005-of-00005.bin",
"base_model.lm_head.weight": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.0.attn.bias": "pytorch_model-00003-of-00005.bin",
"base_model.transformer.h.0.attn.k_proj.weight": "pytorch_model-00003-of-00005.bin",
"base_model.transformer.h.0.attn.masked_bias": "pytorch_model-00003-of-00005.bin",
"base_model.transformer.h.0.attn.out_proj.weight": "pytorch_model-00003-of-00005.bin",
"base_model.transformer.h.0.attn.q_proj.weight": "pytorch_model-00003-of-00005.bin",
"base_model.transformer.h.0.attn.v_proj.weight": "pytorch_model-00003-of-00005.bin",
"base_model.transformer.h.0.ln_1.bias": "pytorch_model-00003-of-00005.bin",
"base_model.transformer.h.0.ln_1.weight": "pytorch_model-00003-of-00005.bin",
"base_model.transformer.h.0.mlp.fc_in.bias": "pytorch_model-00003-of-00005.bin",
"base_model.transformer.h.0.mlp.fc_in.weight": "pytorch_model-00003-of-00005.bin",
"base_model.transformer.h.0.mlp.fc_out.bias": "pytorch_model-00003-of-00005.bin",
"base_model.transformer.h.0.mlp.fc_out.weight": "pytorch_model-00003-of-00005.bin",
"base_model.transformer.h.1.attn.bias": "pytorch_model-00003-of-00005.bin",
"base_model.transformer.h.1.attn.k_proj.weight": "pytorch_model-00003-of-00005.bin",
"base_model.transformer.h.1.attn.masked_bias": "pytorch_model-00003-of-00005.bin",
"base_model.transformer.h.1.attn.out_proj.weight": "pytorch_model-00003-of-00005.bin",
"base_model.transformer.h.1.attn.q_proj.weight": "pytorch_model-00003-of-00005.bin",
"base_model.transformer.h.1.attn.v_proj.weight": "pytorch_model-00003-of-00005.bin",
"base_model.transformer.h.1.ln_1.bias": "pytorch_model-00003-of-00005.bin",
"base_model.transformer.h.1.ln_1.weight": "pytorch_model-00003-of-00005.bin",
"base_model.transformer.h.1.mlp.fc_in.bias": "pytorch_model-00003-of-00005.bin",
"base_model.transformer.h.1.mlp.fc_in.weight": "pytorch_model-00003-of-00005.bin",
"base_model.transformer.h.1.mlp.fc_out.bias": "pytorch_model-00003-of-00005.bin",
"base_model.transformer.h.1.mlp.fc_out.weight": "pytorch_model-00003-of-00005.bin",
"base_model.transformer.h.10.attn.bias": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.10.attn.k_proj.weight": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.10.attn.masked_bias": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.10.attn.out_proj.weight": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.10.attn.q_proj.weight": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.10.attn.v_proj.weight": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.10.ln_1.bias": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.10.ln_1.weight": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.10.mlp.fc_in.bias": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.10.mlp.fc_in.weight": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.10.mlp.fc_out.bias": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.10.mlp.fc_out.weight": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.11.attn.bias": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.11.attn.k_proj.weight": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.11.attn.masked_bias": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.11.attn.out_proj.weight": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.11.attn.q_proj.weight": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.11.attn.v_proj.weight": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.11.ln_1.bias": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.11.ln_1.weight": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.11.mlp.fc_in.bias": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.11.mlp.fc_in.weight": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.11.mlp.fc_out.bias": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.11.mlp.fc_out.weight": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.12.attn.bias": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.12.attn.k_proj.weight": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.12.attn.masked_bias": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.12.attn.out_proj.weight": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.12.attn.q_proj.weight": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.12.attn.v_proj.weight": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.12.ln_1.bias": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.12.ln_1.weight": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.12.mlp.fc_in.bias": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.12.mlp.fc_in.weight": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.12.mlp.fc_out.bias": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.12.mlp.fc_out.weight": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.13.attn.bias": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.13.attn.k_proj.weight": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.13.attn.masked_bias": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.13.attn.out_proj.weight": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.13.attn.q_proj.weight": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.13.attn.v_proj.weight": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.13.ln_1.bias": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.13.ln_1.weight": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.13.mlp.fc_in.bias": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.13.mlp.fc_in.weight": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.13.mlp.fc_out.bias": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.13.mlp.fc_out.weight": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.14.attn.bias": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.14.attn.k_proj.weight": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.14.attn.masked_bias": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.14.attn.out_proj.weight": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.14.attn.q_proj.weight": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.14.attn.v_proj.weight": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.14.ln_1.bias": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.14.ln_1.weight": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.14.mlp.fc_in.bias": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.14.mlp.fc_in.weight": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.14.mlp.fc_out.bias": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.14.mlp.fc_out.weight": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.15.attn.bias": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.15.attn.k_proj.weight": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.15.attn.masked_bias": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.15.attn.out_proj.weight": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.15.attn.q_proj.weight": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.15.attn.v_proj.weight": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.15.ln_1.bias": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.15.ln_1.weight": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.15.mlp.fc_in.bias": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.15.mlp.fc_in.weight": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.15.mlp.fc_out.bias": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.15.mlp.fc_out.weight": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.16.attn.bias": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.16.attn.k_proj.weight": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.16.attn.masked_bias": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.16.attn.out_proj.weight": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.16.attn.q_proj.weight": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.16.attn.v_proj.weight": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.16.ln_1.bias": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.16.ln_1.weight": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.16.mlp.fc_in.bias": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.16.mlp.fc_in.weight": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.16.mlp.fc_out.bias": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.16.mlp.fc_out.weight": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.17.attn.bias": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.17.attn.k_proj.weight": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.17.attn.masked_bias": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.17.attn.out_proj.weight": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.17.attn.q_proj.weight": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.17.attn.v_proj.weight": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.17.ln_1.bias": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.17.ln_1.weight": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.17.mlp.fc_in.bias": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.17.mlp.fc_in.weight": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.17.mlp.fc_out.bias": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.17.mlp.fc_out.weight": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.18.attn.bias": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.18.attn.k_proj.weight": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.18.attn.masked_bias": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.18.attn.out_proj.weight": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.18.attn.q_proj.weight": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.18.attn.v_proj.weight": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.18.ln_1.bias": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.18.ln_1.weight": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.18.mlp.fc_in.bias": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.18.mlp.fc_in.weight": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.18.mlp.fc_out.bias": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.18.mlp.fc_out.weight": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.19.attn.bias": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.19.attn.k_proj.weight": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.19.attn.masked_bias": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.19.attn.out_proj.weight": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.19.attn.q_proj.weight": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.19.attn.v_proj.weight": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.19.ln_1.bias": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.19.ln_1.weight": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.19.mlp.fc_in.bias": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.19.mlp.fc_in.weight": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.19.mlp.fc_out.bias": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.19.mlp.fc_out.weight": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.2.attn.bias": "pytorch_model-00003-of-00005.bin",
"base_model.transformer.h.2.attn.k_proj.weight": "pytorch_model-00003-of-00005.bin",
"base_model.transformer.h.2.attn.masked_bias": "pytorch_model-00003-of-00005.bin",
"base_model.transformer.h.2.attn.out_proj.weight": "pytorch_model-00003-of-00005.bin",
"base_model.transformer.h.2.attn.q_proj.weight": "pytorch_model-00003-of-00005.bin",
"base_model.transformer.h.2.attn.v_proj.weight": "pytorch_model-00003-of-00005.bin",
"base_model.transformer.h.2.ln_1.bias": "pytorch_model-00003-of-00005.bin",
"base_model.transformer.h.2.ln_1.weight": "pytorch_model-00003-of-00005.bin",
"base_model.transformer.h.2.mlp.fc_in.bias": "pytorch_model-00003-of-00005.bin",
"base_model.transformer.h.2.mlp.fc_in.weight": "pytorch_model-00003-of-00005.bin",
"base_model.transformer.h.2.mlp.fc_out.bias": "pytorch_model-00003-of-00005.bin",
"base_model.transformer.h.2.mlp.fc_out.weight": "pytorch_model-00003-of-00005.bin",
"base_model.transformer.h.20.attn.bias": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.20.attn.k_proj.weight": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.20.attn.masked_bias": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.20.attn.out_proj.weight": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.20.attn.q_proj.weight": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.20.attn.v_proj.weight": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.20.ln_1.bias": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.20.ln_1.weight": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.20.mlp.fc_in.bias": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.20.mlp.fc_in.weight": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.20.mlp.fc_out.bias": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.20.mlp.fc_out.weight": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.21.attn.bias": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.21.attn.k_proj.weight": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.21.attn.masked_bias": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.21.attn.out_proj.weight": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.21.attn.q_proj.weight": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.21.attn.v_proj.weight": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.21.ln_1.bias": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.21.ln_1.weight": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.21.mlp.fc_in.bias": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.21.mlp.fc_in.weight": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.21.mlp.fc_out.bias": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.21.mlp.fc_out.weight": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.22.attn.bias": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.22.attn.k_proj.weight": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.22.attn.masked_bias": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.22.attn.out_proj.weight": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.22.attn.q_proj.weight": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.22.attn.v_proj.weight": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.22.ln_1.bias": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.22.ln_1.weight": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.22.mlp.fc_in.bias": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.22.mlp.fc_in.weight": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.22.mlp.fc_out.bias": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.22.mlp.fc_out.weight": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.23.attn.bias": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.23.attn.k_proj.weight": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.23.attn.masked_bias": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.23.attn.out_proj.weight": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.23.attn.q_proj.weight": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.23.attn.v_proj.weight": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.23.ln_1.bias": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.23.ln_1.weight": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.23.mlp.fc_in.bias": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.23.mlp.fc_in.weight": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.23.mlp.fc_out.bias": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.23.mlp.fc_out.weight": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.24.attn.bias": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.24.attn.k_proj.weight": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.24.attn.masked_bias": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.24.attn.out_proj.weight": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.24.attn.q_proj.weight": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.24.attn.v_proj.weight": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.24.ln_1.bias": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.24.ln_1.weight": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.24.mlp.fc_in.bias": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.24.mlp.fc_in.weight": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.24.mlp.fc_out.bias": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.24.mlp.fc_out.weight": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.25.attn.bias": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.25.attn.k_proj.weight": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.25.attn.masked_bias": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.25.attn.out_proj.weight": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.25.attn.q_proj.weight": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.25.attn.v_proj.weight": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.25.ln_1.bias": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.25.ln_1.weight": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.25.mlp.fc_in.bias": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.25.mlp.fc_in.weight": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.25.mlp.fc_out.bias": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.25.mlp.fc_out.weight": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.26.attn.bias": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.26.attn.k_proj.weight": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.26.attn.masked_bias": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.26.attn.out_proj.weight": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.26.attn.q_proj.weight": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.26.attn.v_proj.weight": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.26.ln_1.bias": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.26.ln_1.weight": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.26.mlp.fc_in.bias": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.26.mlp.fc_in.weight": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.26.mlp.fc_out.bias": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.26.mlp.fc_out.weight": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.27.attn.bias": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.27.attn.k_proj.weight": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.27.attn.masked_bias": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.27.attn.out_proj.weight": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.27.attn.q_proj.weight": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.27.attn.v_proj.weight": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.27.ln_1.bias": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.27.ln_1.weight": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.27.mlp.fc_in.bias": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.27.mlp.fc_in.weight": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.27.mlp.fc_out.bias": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.27.mlp.fc_out.weight": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.h.3.attn.bias": "pytorch_model-00003-of-00005.bin",
"base_model.transformer.h.3.attn.k_proj.weight": "pytorch_model-00003-of-00005.bin",
"base_model.transformer.h.3.attn.masked_bias": "pytorch_model-00003-of-00005.bin",
"base_model.transformer.h.3.attn.out_proj.weight": "pytorch_model-00003-of-00005.bin",
"base_model.transformer.h.3.attn.q_proj.weight": "pytorch_model-00003-of-00005.bin",
"base_model.transformer.h.3.attn.v_proj.weight": "pytorch_model-00003-of-00005.bin",
"base_model.transformer.h.3.ln_1.bias": "pytorch_model-00003-of-00005.bin",
"base_model.transformer.h.3.ln_1.weight": "pytorch_model-00003-of-00005.bin",
"base_model.transformer.h.3.mlp.fc_in.bias": "pytorch_model-00003-of-00005.bin",
"base_model.transformer.h.3.mlp.fc_in.weight": "pytorch_model-00003-of-00005.bin",
"base_model.transformer.h.3.mlp.fc_out.bias": "pytorch_model-00003-of-00005.bin",
"base_model.transformer.h.3.mlp.fc_out.weight": "pytorch_model-00003-of-00005.bin",
"base_model.transformer.h.4.attn.bias": "pytorch_model-00003-of-00005.bin",
"base_model.transformer.h.4.attn.k_proj.weight": "pytorch_model-00003-of-00005.bin",
"base_model.transformer.h.4.attn.masked_bias": "pytorch_model-00003-of-00005.bin",
"base_model.transformer.h.4.attn.out_proj.weight": "pytorch_model-00003-of-00005.bin",
"base_model.transformer.h.4.attn.q_proj.weight": "pytorch_model-00003-of-00005.bin",
"base_model.transformer.h.4.attn.v_proj.weight": "pytorch_model-00003-of-00005.bin",
"base_model.transformer.h.4.ln_1.bias": "pytorch_model-00003-of-00005.bin",
"base_model.transformer.h.4.ln_1.weight": "pytorch_model-00003-of-00005.bin",
"base_model.transformer.h.4.mlp.fc_in.bias": "pytorch_model-00003-of-00005.bin",
"base_model.transformer.h.4.mlp.fc_in.weight": "pytorch_model-00003-of-00005.bin",
"base_model.transformer.h.4.mlp.fc_out.bias": "pytorch_model-00003-of-00005.bin",
"base_model.transformer.h.4.mlp.fc_out.weight": "pytorch_model-00003-of-00005.bin",
"base_model.transformer.h.5.attn.bias": "pytorch_model-00003-of-00005.bin",
"base_model.transformer.h.5.attn.k_proj.weight": "pytorch_model-00003-of-00005.bin",
"base_model.transformer.h.5.attn.masked_bias": "pytorch_model-00003-of-00005.bin",
"base_model.transformer.h.5.attn.out_proj.weight": "pytorch_model-00003-of-00005.bin",
"base_model.transformer.h.5.attn.q_proj.weight": "pytorch_model-00003-of-00005.bin",
"base_model.transformer.h.5.attn.v_proj.weight": "pytorch_model-00003-of-00005.bin",
"base_model.transformer.h.5.ln_1.bias": "pytorch_model-00003-of-00005.bin",
"base_model.transformer.h.5.ln_1.weight": "pytorch_model-00003-of-00005.bin",
"base_model.transformer.h.5.mlp.fc_in.bias": "pytorch_model-00003-of-00005.bin",
"base_model.transformer.h.5.mlp.fc_in.weight": "pytorch_model-00003-of-00005.bin",
"base_model.transformer.h.5.mlp.fc_out.bias": "pytorch_model-00003-of-00005.bin",
"base_model.transformer.h.5.mlp.fc_out.weight": "pytorch_model-00003-of-00005.bin",
"base_model.transformer.h.6.attn.bias": "pytorch_model-00003-of-00005.bin",
"base_model.transformer.h.6.attn.k_proj.weight": "pytorch_model-00003-of-00005.bin",
"base_model.transformer.h.6.attn.masked_bias": "pytorch_model-00003-of-00005.bin",
"base_model.transformer.h.6.attn.out_proj.weight": "pytorch_model-00003-of-00005.bin",
"base_model.transformer.h.6.attn.q_proj.weight": "pytorch_model-00003-of-00005.bin",
"base_model.transformer.h.6.attn.v_proj.weight": "pytorch_model-00003-of-00005.bin",
"base_model.transformer.h.6.ln_1.bias": "pytorch_model-00003-of-00005.bin",
"base_model.transformer.h.6.ln_1.weight": "pytorch_model-00003-of-00005.bin",
"base_model.transformer.h.6.mlp.fc_in.bias": "pytorch_model-00003-of-00005.bin",
"base_model.transformer.h.6.mlp.fc_in.weight": "pytorch_model-00003-of-00005.bin",
"base_model.transformer.h.6.mlp.fc_out.bias": "pytorch_model-00003-of-00005.bin",
"base_model.transformer.h.6.mlp.fc_out.weight": "pytorch_model-00003-of-00005.bin",
"base_model.transformer.h.7.attn.bias": "pytorch_model-00003-of-00005.bin",
"base_model.transformer.h.7.attn.k_proj.weight": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.7.attn.masked_bias": "pytorch_model-00003-of-00005.bin",
"base_model.transformer.h.7.attn.out_proj.weight": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.7.attn.q_proj.weight": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.7.attn.v_proj.weight": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.7.ln_1.bias": "pytorch_model-00003-of-00005.bin",
"base_model.transformer.h.7.ln_1.weight": "pytorch_model-00003-of-00005.bin",
"base_model.transformer.h.7.mlp.fc_in.bias": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.7.mlp.fc_in.weight": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.7.mlp.fc_out.bias": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.7.mlp.fc_out.weight": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.8.attn.bias": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.8.attn.k_proj.weight": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.8.attn.masked_bias": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.8.attn.out_proj.weight": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.8.attn.q_proj.weight": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.8.attn.v_proj.weight": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.8.ln_1.bias": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.8.ln_1.weight": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.8.mlp.fc_in.bias": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.8.mlp.fc_in.weight": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.8.mlp.fc_out.bias": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.8.mlp.fc_out.weight": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.9.attn.bias": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.9.attn.k_proj.weight": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.9.attn.masked_bias": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.9.attn.out_proj.weight": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.9.attn.q_proj.weight": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.9.attn.v_proj.weight": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.9.ln_1.bias": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.9.ln_1.weight": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.9.mlp.fc_in.bias": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.9.mlp.fc_in.weight": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.9.mlp.fc_out.bias": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.h.9.mlp.fc_out.weight": "pytorch_model-00004-of-00005.bin",
"base_model.transformer.ln_f.bias": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.ln_f.weight": "pytorch_model-00005-of-00005.bin",
"base_model.transformer.wte.weight": "pytorch_model-00003-of-00005.bin",
"transformer.h.0.attn.bias": "pytorch_model-00001-of-00005.bin",
"transformer.h.0.attn.k_proj.weight": "pytorch_model-00001-of-00005.bin",
"transformer.h.0.attn.masked_bias": "pytorch_model-00001-of-00005.bin",
"transformer.h.0.attn.out_proj.weight": "pytorch_model-00001-of-00005.bin",
"transformer.h.0.attn.q_proj.weight": "pytorch_model-00001-of-00005.bin",
"transformer.h.0.attn.v_proj.weight": "pytorch_model-00001-of-00005.bin",
"transformer.h.0.ln_1.bias": "pytorch_model-00001-of-00005.bin",
"transformer.h.0.ln_1.weight": "pytorch_model-00001-of-00005.bin",
"transformer.h.0.mlp.fc_in.bias": "pytorch_model-00001-of-00005.bin",
"transformer.h.0.mlp.fc_in.weight": "pytorch_model-00001-of-00005.bin",
"transformer.h.0.mlp.fc_out.bias": "pytorch_model-00001-of-00005.bin",
"transformer.h.0.mlp.fc_out.weight": "pytorch_model-00001-of-00005.bin",
"transformer.h.1.attn.bias": "pytorch_model-00001-of-00005.bin",
"transformer.h.1.attn.k_proj.weight": "pytorch_model-00001-of-00005.bin",
"transformer.h.1.attn.masked_bias": "pytorch_model-00001-of-00005.bin",
"transformer.h.1.attn.out_proj.weight": "pytorch_model-00001-of-00005.bin",
"transformer.h.1.attn.q_proj.weight": "pytorch_model-00001-of-00005.bin",
"transformer.h.1.attn.v_proj.weight": "pytorch_model-00001-of-00005.bin",
"transformer.h.1.ln_1.bias": "pytorch_model-00001-of-00005.bin",
"transformer.h.1.ln_1.weight": "pytorch_model-00001-of-00005.bin",
"transformer.h.1.mlp.fc_in.bias": "pytorch_model-00001-of-00005.bin",
"transformer.h.1.mlp.fc_in.weight": "pytorch_model-00001-of-00005.bin",
"transformer.h.1.mlp.fc_out.bias": "pytorch_model-00001-of-00005.bin",
"transformer.h.1.mlp.fc_out.weight": "pytorch_model-00001-of-00005.bin",
"transformer.h.10.attn.bias": "pytorch_model-00001-of-00005.bin",
"transformer.h.10.attn.k_proj.weight": "pytorch_model-00001-of-00005.bin",
"transformer.h.10.attn.masked_bias": "pytorch_model-00001-of-00005.bin",
"transformer.h.10.attn.out_proj.weight": "pytorch_model-00001-of-00005.bin",
"transformer.h.10.attn.q_proj.weight": "pytorch_model-00001-of-00005.bin",
"transformer.h.10.attn.v_proj.weight": "pytorch_model-00001-of-00005.bin",
"transformer.h.10.ln_1.bias": "pytorch_model-00001-of-00005.bin",
"transformer.h.10.ln_1.weight": "pytorch_model-00001-of-00005.bin",
"transformer.h.10.mlp.fc_in.bias": "pytorch_model-00001-of-00005.bin",
"transformer.h.10.mlp.fc_in.weight": "pytorch_model-00001-of-00005.bin",
"transformer.h.10.mlp.fc_out.bias": "pytorch_model-00001-of-00005.bin",
"transformer.h.10.mlp.fc_out.weight": "pytorch_model-00001-of-00005.bin",
"transformer.h.11.attn.bias": "pytorch_model-00001-of-00005.bin",
"transformer.h.11.attn.k_proj.weight": "pytorch_model-00001-of-00005.bin",
"transformer.h.11.attn.masked_bias": "pytorch_model-00001-of-00005.bin",
"transformer.h.11.attn.out_proj.weight": "pytorch_model-00001-of-00005.bin",
"transformer.h.11.attn.q_proj.weight": "pytorch_model-00001-of-00005.bin",
"transformer.h.11.attn.v_proj.weight": "pytorch_model-00001-of-00005.bin",
"transformer.h.11.ln_1.bias": "pytorch_model-00001-of-00005.bin",
"transformer.h.11.ln_1.weight": "pytorch_model-00001-of-00005.bin",
"transformer.h.11.mlp.fc_in.bias": "pytorch_model-00002-of-00005.bin",
"transformer.h.11.mlp.fc_in.weight": "pytorch_model-00002-of-00005.bin",
"transformer.h.11.mlp.fc_out.bias": "pytorch_model-00002-of-00005.bin",
"transformer.h.11.mlp.fc_out.weight": "pytorch_model-00002-of-00005.bin",
"transformer.h.12.attn.bias": "pytorch_model-00002-of-00005.bin",
"transformer.h.12.attn.k_proj.weight": "pytorch_model-00002-of-00005.bin",
"transformer.h.12.attn.masked_bias": "pytorch_model-00002-of-00005.bin",
"transformer.h.12.attn.out_proj.weight": "pytorch_model-00002-of-00005.bin",
"transformer.h.12.attn.q_proj.weight": "pytorch_model-00002-of-00005.bin",
"transformer.h.12.attn.v_proj.weight": "pytorch_model-00002-of-00005.bin",
"transformer.h.12.ln_1.bias": "pytorch_model-00002-of-00005.bin",
"transformer.h.12.ln_1.weight": "pytorch_model-00002-of-00005.bin",
"transformer.h.12.mlp.fc_in.bias": "pytorch_model-00002-of-00005.bin",
"transformer.h.12.mlp.fc_in.weight": "pytorch_model-00002-of-00005.bin",
"transformer.h.12.mlp.fc_out.bias": "pytorch_model-00002-of-00005.bin",
"transformer.h.12.mlp.fc_out.weight": "pytorch_model-00002-of-00005.bin",
"transformer.h.13.attn.bias": "pytorch_model-00002-of-00005.bin",
"transformer.h.13.attn.k_proj.weight": "pytorch_model-00002-of-00005.bin",
"transformer.h.13.attn.masked_bias": "pytorch_model-00002-of-00005.bin",
"transformer.h.13.attn.out_proj.weight": "pytorch_model-00002-of-00005.bin",
"transformer.h.13.attn.q_proj.weight": "pytorch_model-00002-of-00005.bin",
"transformer.h.13.attn.v_proj.weight": "pytorch_model-00002-of-00005.bin",
"transformer.h.13.ln_1.bias": "pytorch_model-00002-of-00005.bin",
"transformer.h.13.ln_1.weight": "pytorch_model-00002-of-00005.bin",
"transformer.h.13.mlp.fc_in.bias": "pytorch_model-00002-of-00005.bin",
"transformer.h.13.mlp.fc_in.weight": "pytorch_model-00002-of-00005.bin",
"transformer.h.13.mlp.fc_out.bias": "pytorch_model-00002-of-00005.bin",
"transformer.h.13.mlp.fc_out.weight": "pytorch_model-00002-of-00005.bin",
"transformer.h.14.attn.bias": "pytorch_model-00002-of-00005.bin",
"transformer.h.14.attn.k_proj.weight": "pytorch_model-00002-of-00005.bin",
"transformer.h.14.attn.masked_bias": "pytorch_model-00002-of-00005.bin",
"transformer.h.14.attn.out_proj.weight": "pytorch_model-00002-of-00005.bin",
"transformer.h.14.attn.q_proj.weight": "pytorch_model-00002-of-00005.bin",
"transformer.h.14.attn.v_proj.weight": "pytorch_model-00002-of-00005.bin",
"transformer.h.14.ln_1.bias": "pytorch_model-00002-of-00005.bin",
"transformer.h.14.ln_1.weight": "pytorch_model-00002-of-00005.bin",
"transformer.h.14.mlp.fc_in.bias": "pytorch_model-00002-of-00005.bin",
"transformer.h.14.mlp.fc_in.weight": "pytorch_model-00002-of-00005.bin",
"transformer.h.14.mlp.fc_out.bias": "pytorch_model-00002-of-00005.bin",
"transformer.h.14.mlp.fc_out.weight": "pytorch_model-00002-of-00005.bin",
"transformer.h.15.attn.bias": "pytorch_model-00002-of-00005.bin",
"transformer.h.15.attn.k_proj.weight": "pytorch_model-00002-of-00005.bin",
"transformer.h.15.attn.masked_bias": "pytorch_model-00002-of-00005.bin",
"transformer.h.15.attn.out_proj.weight": "pytorch_model-00002-of-00005.bin",
"transformer.h.15.attn.q_proj.weight": "pytorch_model-00002-of-00005.bin",
"transformer.h.15.attn.v_proj.weight": "pytorch_model-00002-of-00005.bin",
"transformer.h.15.ln_1.bias": "pytorch_model-00002-of-00005.bin",
"transformer.h.15.ln_1.weight": "pytorch_model-00002-of-00005.bin",
"transformer.h.15.mlp.fc_in.bias": "pytorch_model-00002-of-00005.bin",
"transformer.h.15.mlp.fc_in.weight": "pytorch_model-00002-of-00005.bin",
"transformer.h.15.mlp.fc_out.bias": "pytorch_model-00002-of-00005.bin",
"transformer.h.15.mlp.fc_out.weight": "pytorch_model-00002-of-00005.bin",
"transformer.h.16.attn.bias": "pytorch_model-00002-of-00005.bin",
"transformer.h.16.attn.k_proj.weight": "pytorch_model-00002-of-00005.bin",
"transformer.h.16.attn.masked_bias": "pytorch_model-00002-of-00005.bin",
"transformer.h.16.attn.out_proj.weight": "pytorch_model-00002-of-00005.bin",
"transformer.h.16.attn.q_proj.weight": "pytorch_model-00002-of-00005.bin",
"transformer.h.16.attn.v_proj.weight": "pytorch_model-00002-of-00005.bin",
"transformer.h.16.ln_1.bias": "pytorch_model-00002-of-00005.bin",
"transformer.h.16.ln_1.weight": "pytorch_model-00002-of-00005.bin",
"transformer.h.16.mlp.fc_in.bias": "pytorch_model-00002-of-00005.bin",
"transformer.h.16.mlp.fc_in.weight": "pytorch_model-00002-of-00005.bin",
"transformer.h.16.mlp.fc_out.bias": "pytorch_model-00002-of-00005.bin",
"transformer.h.16.mlp.fc_out.weight": "pytorch_model-00002-of-00005.bin",
"transformer.h.17.attn.bias": "pytorch_model-00002-of-00005.bin",
"transformer.h.17.attn.k_proj.weight": "pytorch_model-00002-of-00005.bin",
"transformer.h.17.attn.masked_bias": "pytorch_model-00002-of-00005.bin",
"transformer.h.17.attn.out_proj.weight": "pytorch_model-00002-of-00005.bin",
"transformer.h.17.attn.q_proj.weight": "pytorch_model-00002-of-00005.bin",
"transformer.h.17.attn.v_proj.weight": "pytorch_model-00002-of-00005.bin",
"transformer.h.17.ln_1.bias": "pytorch_model-00002-of-00005.bin",
"transformer.h.17.ln_1.weight": "pytorch_model-00002-of-00005.bin",
"transformer.h.17.mlp.fc_in.bias": "pytorch_model-00002-of-00005.bin",
"transformer.h.17.mlp.fc_in.weight": "pytorch_model-00002-of-00005.bin",
"transformer.h.17.mlp.fc_out.bias": "pytorch_model-00002-of-00005.bin",
"transformer.h.17.mlp.fc_out.weight": "pytorch_model-00002-of-00005.bin",
"transformer.h.18.attn.bias": "pytorch_model-00002-of-00005.bin",
"transformer.h.18.attn.k_proj.weight": "pytorch_model-00002-of-00005.bin",
"transformer.h.18.attn.masked_bias": "pytorch_model-00002-of-00005.bin",
"transformer.h.18.attn.out_proj.weight": "pytorch_model-00002-of-00005.bin",
"transformer.h.18.attn.q_proj.weight": "pytorch_model-00002-of-00005.bin",
"transformer.h.18.attn.v_proj.weight": "pytorch_model-00002-of-00005.bin",
"transformer.h.18.ln_1.bias": "pytorch_model-00002-of-00005.bin",
"transformer.h.18.ln_1.weight": "pytorch_model-00002-of-00005.bin",
"transformer.h.18.mlp.fc_in.bias": "pytorch_model-00002-of-00005.bin",
"transformer.h.18.mlp.fc_in.weight": "pytorch_model-00002-of-00005.bin",
"transformer.h.18.mlp.fc_out.bias": "pytorch_model-00002-of-00005.bin",
"transformer.h.18.mlp.fc_out.weight": "pytorch_model-00002-of-00005.bin",
"transformer.h.19.attn.bias": "pytorch_model-00002-of-00005.bin",
"transformer.h.19.attn.k_proj.weight": "pytorch_model-00002-of-00005.bin",
"transformer.h.19.attn.masked_bias": "pytorch_model-00002-of-00005.bin",
"transformer.h.19.attn.out_proj.weight": "pytorch_model-00002-of-00005.bin",
"transformer.h.19.attn.q_proj.weight": "pytorch_model-00002-of-00005.bin",
"transformer.h.19.attn.v_proj.weight": "pytorch_model-00002-of-00005.bin",
"transformer.h.19.ln_1.bias": "pytorch_model-00002-of-00005.bin",
"transformer.h.19.ln_1.weight": "pytorch_model-00002-of-00005.bin",
"transformer.h.19.mlp.fc_in.bias": "pytorch_model-00002-of-00005.bin",
"transformer.h.19.mlp.fc_in.weight": "pytorch_model-00002-of-00005.bin",
"transformer.h.19.mlp.fc_out.bias": "pytorch_model-00002-of-00005.bin",
"transformer.h.19.mlp.fc_out.weight": "pytorch_model-00002-of-00005.bin",
"transformer.h.2.attn.bias": "pytorch_model-00001-of-00005.bin",
"transformer.h.2.attn.k_proj.weight": "pytorch_model-00001-of-00005.bin",
"transformer.h.2.attn.masked_bias": "pytorch_model-00001-of-00005.bin",
"transformer.h.2.attn.out_proj.weight": "pytorch_model-00001-of-00005.bin",
"transformer.h.2.attn.q_proj.weight": "pytorch_model-00001-of-00005.bin",
"transformer.h.2.attn.v_proj.weight": "pytorch_model-00001-of-00005.bin",
"transformer.h.2.ln_1.bias": "pytorch_model-00001-of-00005.bin",
"transformer.h.2.ln_1.weight": "pytorch_model-00001-of-00005.bin",
"transformer.h.2.mlp.fc_in.bias": "pytorch_model-00001-of-00005.bin",
"transformer.h.2.mlp.fc_in.weight": "pytorch_model-00001-of-00005.bin",
"transformer.h.2.mlp.fc_out.bias": "pytorch_model-00001-of-00005.bin",
"transformer.h.2.mlp.fc_out.weight": "pytorch_model-00001-of-00005.bin",
"transformer.h.20.attn.bias": "pytorch_model-00002-of-00005.bin",
"transformer.h.20.attn.k_proj.weight": "pytorch_model-00002-of-00005.bin",
"transformer.h.20.attn.masked_bias": "pytorch_model-00002-of-00005.bin",
"transformer.h.20.attn.out_proj.weight": "pytorch_model-00002-of-00005.bin",
"transformer.h.20.attn.q_proj.weight": "pytorch_model-00002-of-00005.bin",
"transformer.h.20.attn.v_proj.weight": "pytorch_model-00002-of-00005.bin",
"transformer.h.20.ln_1.bias": "pytorch_model-00002-of-00005.bin",
"transformer.h.20.ln_1.weight": "pytorch_model-00002-of-00005.bin",
"transformer.h.20.mlp.fc_in.bias": "pytorch_model-00002-of-00005.bin",
"transformer.h.20.mlp.fc_in.weight": "pytorch_model-00002-of-00005.bin",
"transformer.h.20.mlp.fc_out.bias": "pytorch_model-00002-of-00005.bin",
"transformer.h.20.mlp.fc_out.weight": "pytorch_model-00002-of-00005.bin",
"transformer.h.21.attn.bias": "pytorch_model-00002-of-00005.bin",
"transformer.h.21.attn.k_proj.weight": "pytorch_model-00002-of-00005.bin",
"transformer.h.21.attn.masked_bias": "pytorch_model-00002-of-00005.bin",
"transformer.h.21.attn.out_proj.weight": "pytorch_model-00002-of-00005.bin",
"transformer.h.21.attn.q_proj.weight": "pytorch_model-00002-of-00005.bin",
"transformer.h.21.attn.v_proj.weight": "pytorch_model-00002-of-00005.bin",
"transformer.h.21.ln_1.bias": "pytorch_model-00002-of-00005.bin",
"transformer.h.21.ln_1.weight": "pytorch_model-00002-of-00005.bin",
"transformer.h.21.mlp.fc_in.bias": "pytorch_model-00002-of-00005.bin",
"transformer.h.21.mlp.fc_in.weight": "pytorch_model-00002-of-00005.bin",
"transformer.h.21.mlp.fc_out.bias": "pytorch_model-00002-of-00005.bin",
"transformer.h.21.mlp.fc_out.weight": "pytorch_model-00002-of-00005.bin",
"transformer.h.22.attn.bias": "pytorch_model-00002-of-00005.bin",
"transformer.h.22.attn.k_proj.weight": "pytorch_model-00002-of-00005.bin",
"transformer.h.22.attn.masked_bias": "pytorch_model-00002-of-00005.bin",
"transformer.h.22.attn.out_proj.weight": "pytorch_model-00002-of-00005.bin",
"transformer.h.22.attn.q_proj.weight": "pytorch_model-00002-of-00005.bin",
"transformer.h.22.attn.v_proj.weight": "pytorch_model-00002-of-00005.bin",
"transformer.h.22.ln_1.bias": "pytorch_model-00002-of-00005.bin",
"transformer.h.22.ln_1.weight": "pytorch_model-00002-of-00005.bin",
"transformer.h.22.mlp.fc_in.bias": "pytorch_model-00002-of-00005.bin",
"transformer.h.22.mlp.fc_in.weight": "pytorch_model-00002-of-00005.bin",
"transformer.h.22.mlp.fc_out.bias": "pytorch_model-00002-of-00005.bin",
"transformer.h.22.mlp.fc_out.weight": "pytorch_model-00002-of-00005.bin",
"transformer.h.23.attn.bias": "pytorch_model-00002-of-00005.bin",
"transformer.h.23.attn.k_proj.weight": "pytorch_model-00002-of-00005.bin",
"transformer.h.23.attn.masked_bias": "pytorch_model-00002-of-00005.bin",
"transformer.h.23.attn.out_proj.weight": "pytorch_model-00002-of-00005.bin",
"transformer.h.23.attn.q_proj.weight": "pytorch_model-00002-of-00005.bin",
"transformer.h.23.attn.v_proj.weight": "pytorch_model-00002-of-00005.bin",
"transformer.h.23.ln_1.bias": "pytorch_model-00002-of-00005.bin",
"transformer.h.23.ln_1.weight": "pytorch_model-00002-of-00005.bin",
"transformer.h.23.mlp.fc_in.bias": "pytorch_model-00002-of-00005.bin",
"transformer.h.23.mlp.fc_in.weight": "pytorch_model-00002-of-00005.bin",
"transformer.h.23.mlp.fc_out.bias": "pytorch_model-00003-of-00005.bin",
"transformer.h.23.mlp.fc_out.weight": "pytorch_model-00003-of-00005.bin",
"transformer.h.24.attn.bias": "pytorch_model-00003-of-00005.bin",
"transformer.h.24.attn.k_proj.weight": "pytorch_model-00003-of-00005.bin",
"transformer.h.24.attn.masked_bias": "pytorch_model-00003-of-00005.bin",
"transformer.h.24.attn.out_proj.weight": "pytorch_model-00003-of-00005.bin",
"transformer.h.24.attn.q_proj.weight": "pytorch_model-00003-of-00005.bin",
"transformer.h.24.attn.v_proj.weight": "pytorch_model-00003-of-00005.bin",
"transformer.h.24.ln_1.bias": "pytorch_model-00003-of-00005.bin",
"transformer.h.24.ln_1.weight": "pytorch_model-00003-of-00005.bin",
"transformer.h.24.mlp.fc_in.bias": "pytorch_model-00003-of-00005.bin",
"transformer.h.24.mlp.fc_in.weight": "pytorch_model-00003-of-00005.bin",
"transformer.h.24.mlp.fc_out.bias": "pytorch_model-00003-of-00005.bin",
"transformer.h.24.mlp.fc_out.weight": "pytorch_model-00003-of-00005.bin",
"transformer.h.25.attn.bias": "pytorch_model-00003-of-00005.bin",
"transformer.h.25.attn.k_proj.weight": "pytorch_model-00003-of-00005.bin",
"transformer.h.25.attn.masked_bias": "pytorch_model-00003-of-00005.bin",
"transformer.h.25.attn.out_proj.weight": "pytorch_model-00003-of-00005.bin",
"transformer.h.25.attn.q_proj.weight": "pytorch_model-00003-of-00005.bin",
"transformer.h.25.attn.v_proj.weight": "pytorch_model-00003-of-00005.bin",
"transformer.h.25.ln_1.bias": "pytorch_model-00003-of-00005.bin",
"transformer.h.25.ln_1.weight": "pytorch_model-00003-of-00005.bin",
"transformer.h.25.mlp.fc_in.bias": "pytorch_model-00003-of-00005.bin",
"transformer.h.25.mlp.fc_in.weight": "pytorch_model-00003-of-00005.bin",
"transformer.h.25.mlp.fc_out.bias": "pytorch_model-00003-of-00005.bin",
"transformer.h.25.mlp.fc_out.weight": "pytorch_model-00003-of-00005.bin",
"transformer.h.26.attn.bias": "pytorch_model-00003-of-00005.bin",
"transformer.h.26.attn.k_proj.weight": "pytorch_model-00003-of-00005.bin",
"transformer.h.26.attn.masked_bias": "pytorch_model-00003-of-00005.bin",
"transformer.h.26.attn.out_proj.weight": "pytorch_model-00003-of-00005.bin",
"transformer.h.26.attn.q_proj.weight": "pytorch_model-00003-of-00005.bin",
"transformer.h.26.attn.v_proj.weight": "pytorch_model-00003-of-00005.bin",
"transformer.h.26.ln_1.bias": "pytorch_model-00003-of-00005.bin",
"transformer.h.26.ln_1.weight": "pytorch_model-00003-of-00005.bin",
"transformer.h.26.mlp.fc_in.bias": "pytorch_model-00003-of-00005.bin",
"transformer.h.26.mlp.fc_in.weight": "pytorch_model-00003-of-00005.bin",
"transformer.h.26.mlp.fc_out.bias": "pytorch_model-00003-of-00005.bin",
"transformer.h.26.mlp.fc_out.weight": "pytorch_model-00003-of-00005.bin",
"transformer.h.27.attn.bias": "pytorch_model-00003-of-00005.bin",
"transformer.h.27.attn.k_proj.weight": "pytorch_model-00003-of-00005.bin",
"transformer.h.27.attn.masked_bias": "pytorch_model-00003-of-00005.bin",
"transformer.h.27.attn.out_proj.weight": "pytorch_model-00003-of-00005.bin",
"transformer.h.27.attn.q_proj.weight": "pytorch_model-00003-of-00005.bin",
"transformer.h.27.attn.v_proj.weight": "pytorch_model-00003-of-00005.bin",
"transformer.h.27.ln_1.bias": "pytorch_model-00003-of-00005.bin",
"transformer.h.27.ln_1.weight": "pytorch_model-00003-of-00005.bin",
"transformer.h.27.mlp.fc_in.bias": "pytorch_model-00003-of-00005.bin",
"transformer.h.27.mlp.fc_in.weight": "pytorch_model-00003-of-00005.bin",
"transformer.h.27.mlp.fc_out.bias": "pytorch_model-00003-of-00005.bin",
"transformer.h.27.mlp.fc_out.weight": "pytorch_model-00003-of-00005.bin",
"transformer.h.3.attn.bias": "pytorch_model-00001-of-00005.bin",
"transformer.h.3.attn.k_proj.weight": "pytorch_model-00001-of-00005.bin",
"transformer.h.3.attn.masked_bias": "pytorch_model-00001-of-00005.bin",
"transformer.h.3.attn.out_proj.weight": "pytorch_model-00001-of-00005.bin",
"transformer.h.3.attn.q_proj.weight": "pytorch_model-00001-of-00005.bin",
"transformer.h.3.attn.v_proj.weight": "pytorch_model-00001-of-00005.bin",
"transformer.h.3.ln_1.bias": "pytorch_model-00001-of-00005.bin",
"transformer.h.3.ln_1.weight": "pytorch_model-00001-of-00005.bin",
"transformer.h.3.mlp.fc_in.bias": "pytorch_model-00001-of-00005.bin",
"transformer.h.3.mlp.fc_in.weight": "pytorch_model-00001-of-00005.bin",
"transformer.h.3.mlp.fc_out.bias": "pytorch_model-00001-of-00005.bin",
"transformer.h.3.mlp.fc_out.weight": "pytorch_model-00001-of-00005.bin",
"transformer.h.4.attn.bias": "pytorch_model-00001-of-00005.bin",
"transformer.h.4.attn.k_proj.weight": "pytorch_model-00001-of-00005.bin",
"transformer.h.4.attn.masked_bias": "pytorch_model-00001-of-00005.bin",
"transformer.h.4.attn.out_proj.weight": "pytorch_model-00001-of-00005.bin",
"transformer.h.4.attn.q_proj.weight": "pytorch_model-00001-of-00005.bin",
"transformer.h.4.attn.v_proj.weight": "pytorch_model-00001-of-00005.bin",
"transformer.h.4.ln_1.bias": "pytorch_model-00001-of-00005.bin",
"transformer.h.4.ln_1.weight": "pytorch_model-00001-of-00005.bin",
"transformer.h.4.mlp.fc_in.bias": "pytorch_model-00001-of-00005.bin",
"transformer.h.4.mlp.fc_in.weight": "pytorch_model-00001-of-00005.bin",
"transformer.h.4.mlp.fc_out.bias": "pytorch_model-00001-of-00005.bin",
"transformer.h.4.mlp.fc_out.weight": "pytorch_model-00001-of-00005.bin",
"transformer.h.5.attn.bias": "pytorch_model-00001-of-00005.bin",
"transformer.h.5.attn.k_proj.weight": "pytorch_model-00001-of-00005.bin",
"transformer.h.5.attn.masked_bias": "pytorch_model-00001-of-00005.bin",
"transformer.h.5.attn.out_proj.weight": "pytorch_model-00001-of-00005.bin",
"transformer.h.5.attn.q_proj.weight": "pytorch_model-00001-of-00005.bin",
"transformer.h.5.attn.v_proj.weight": "pytorch_model-00001-of-00005.bin",
"transformer.h.5.ln_1.bias": "pytorch_model-00001-of-00005.bin",
"transformer.h.5.ln_1.weight": "pytorch_model-00001-of-00005.bin",
"transformer.h.5.mlp.fc_in.bias": "pytorch_model-00001-of-00005.bin",
"transformer.h.5.mlp.fc_in.weight": "pytorch_model-00001-of-00005.bin",
"transformer.h.5.mlp.fc_out.bias": "pytorch_model-00001-of-00005.bin",
"transformer.h.5.mlp.fc_out.weight": "pytorch_model-00001-of-00005.bin",
"transformer.h.6.attn.bias": "pytorch_model-00001-of-00005.bin",
"transformer.h.6.attn.k_proj.weight": "pytorch_model-00001-of-00005.bin",
"transformer.h.6.attn.masked_bias": "pytorch_model-00001-of-00005.bin",
"transformer.h.6.attn.out_proj.weight": "pytorch_model-00001-of-00005.bin",
"transformer.h.6.attn.q_proj.weight": "pytorch_model-00001-of-00005.bin",
"transformer.h.6.attn.v_proj.weight": "pytorch_model-00001-of-00005.bin",
"transformer.h.6.ln_1.bias": "pytorch_model-00001-of-00005.bin",
"transformer.h.6.ln_1.weight": "pytorch_model-00001-of-00005.bin",
"transformer.h.6.mlp.fc_in.bias": "pytorch_model-00001-of-00005.bin",
"transformer.h.6.mlp.fc_in.weight": "pytorch_model-00001-of-00005.bin",
"transformer.h.6.mlp.fc_out.bias": "pytorch_model-00001-of-00005.bin",
"transformer.h.6.mlp.fc_out.weight": "pytorch_model-00001-of-00005.bin",
"transformer.h.7.attn.bias": "pytorch_model-00001-of-00005.bin",
"transformer.h.7.attn.k_proj.weight": "pytorch_model-00001-of-00005.bin",
"transformer.h.7.attn.masked_bias": "pytorch_model-00001-of-00005.bin",
"transformer.h.7.attn.out_proj.weight": "pytorch_model-00001-of-00005.bin",
"transformer.h.7.attn.q_proj.weight": "pytorch_model-00001-of-00005.bin",
"transformer.h.7.attn.v_proj.weight": "pytorch_model-00001-of-00005.bin",
"transformer.h.7.ln_1.bias": "pytorch_model-00001-of-00005.bin",
"transformer.h.7.ln_1.weight": "pytorch_model-00001-of-00005.bin",
"transformer.h.7.mlp.fc_in.bias": "pytorch_model-00001-of-00005.bin",
"transformer.h.7.mlp.fc_in.weight": "pytorch_model-00001-of-00005.bin",
"transformer.h.7.mlp.fc_out.bias": "pytorch_model-00001-of-00005.bin",
"transformer.h.7.mlp.fc_out.weight": "pytorch_model-00001-of-00005.bin",
"transformer.h.8.attn.bias": "pytorch_model-00001-of-00005.bin",
"transformer.h.8.attn.k_proj.weight": "pytorch_model-00001-of-00005.bin",
"transformer.h.8.attn.masked_bias": "pytorch_model-00001-of-00005.bin",
"transformer.h.8.attn.out_proj.weight": "pytorch_model-00001-of-00005.bin",
"transformer.h.8.attn.q_proj.weight": "pytorch_model-00001-of-00005.bin",
"transformer.h.8.attn.v_proj.weight": "pytorch_model-00001-of-00005.bin",
"transformer.h.8.ln_1.bias": "pytorch_model-00001-of-00005.bin",
"transformer.h.8.ln_1.weight": "pytorch_model-00001-of-00005.bin",
"transformer.h.8.mlp.fc_in.bias": "pytorch_model-00001-of-00005.bin",
"transformer.h.8.mlp.fc_in.weight": "pytorch_model-00001-of-00005.bin",
"transformer.h.8.mlp.fc_out.bias": "pytorch_model-00001-of-00005.bin",
"transformer.h.8.mlp.fc_out.weight": "pytorch_model-00001-of-00005.bin",
"transformer.h.9.attn.bias": "pytorch_model-00001-of-00005.bin",
"transformer.h.9.attn.k_proj.weight": "pytorch_model-00001-of-00005.bin",
"transformer.h.9.attn.masked_bias": "pytorch_model-00001-of-00005.bin",
"transformer.h.9.attn.out_proj.weight": "pytorch_model-00001-of-00005.bin",
"transformer.h.9.attn.q_proj.weight": "pytorch_model-00001-of-00005.bin",
"transformer.h.9.attn.v_proj.weight": "pytorch_model-00001-of-00005.bin",
"transformer.h.9.ln_1.bias": "pytorch_model-00001-of-00005.bin",
"transformer.h.9.ln_1.weight": "pytorch_model-00001-of-00005.bin",
"transformer.h.9.mlp.fc_in.bias": "pytorch_model-00001-of-00005.bin",
"transformer.h.9.mlp.fc_in.weight": "pytorch_model-00001-of-00005.bin",
"transformer.h.9.mlp.fc_out.bias": "pytorch_model-00001-of-00005.bin",
"transformer.h.9.mlp.fc_out.weight": "pytorch_model-00001-of-00005.bin",
"transformer.ln_f.bias": "pytorch_model-00003-of-00005.bin",
"transformer.ln_f.weight": "pytorch_model-00003-of-00005.bin",
"transformer.wte.weight": "pytorch_model-00001-of-00005.bin",
"v_head.weight": "pytorch_model-00003-of-00005.bin"
}
}