phi-2 / pytorch_model.bin.index.json
adarshxs's picture
Upload fi-2
1ebb610
{
"metadata": {
"total_size": 11118735360
},
"weight_map": {
"layers.0.wte.weight": "pytorch_model-00001-of-00002.bin",
"layers.1.ln.bias": "pytorch_model-00001-of-00002.bin",
"layers.1.ln.weight": "pytorch_model-00001-of-00002.bin",
"layers.1.mixer.Wqkv.bias": "pytorch_model-00001-of-00002.bin",
"layers.1.mixer.Wqkv.weight": "pytorch_model-00001-of-00002.bin",
"layers.1.mixer.out_proj.bias": "pytorch_model-00001-of-00002.bin",
"layers.1.mixer.out_proj.weight": "pytorch_model-00001-of-00002.bin",
"layers.1.mlp.fc1.bias": "pytorch_model-00001-of-00002.bin",
"layers.1.mlp.fc1.weight": "pytorch_model-00001-of-00002.bin",
"layers.1.mlp.fc2.bias": "pytorch_model-00001-of-00002.bin",
"layers.1.mlp.fc2.weight": "pytorch_model-00001-of-00002.bin",
"layers.10.ln.bias": "pytorch_model-00001-of-00002.bin",
"layers.10.ln.weight": "pytorch_model-00001-of-00002.bin",
"layers.10.mixer.Wqkv.bias": "pytorch_model-00001-of-00002.bin",
"layers.10.mixer.Wqkv.weight": "pytorch_model-00001-of-00002.bin",
"layers.10.mixer.out_proj.bias": "pytorch_model-00001-of-00002.bin",
"layers.10.mixer.out_proj.weight": "pytorch_model-00001-of-00002.bin",
"layers.10.mlp.fc1.bias": "pytorch_model-00001-of-00002.bin",
"layers.10.mlp.fc1.weight": "pytorch_model-00001-of-00002.bin",
"layers.10.mlp.fc2.bias": "pytorch_model-00001-of-00002.bin",
"layers.10.mlp.fc2.weight": "pytorch_model-00001-of-00002.bin",
"layers.11.ln.bias": "pytorch_model-00001-of-00002.bin",
"layers.11.ln.weight": "pytorch_model-00001-of-00002.bin",
"layers.11.mixer.Wqkv.bias": "pytorch_model-00001-of-00002.bin",
"layers.11.mixer.Wqkv.weight": "pytorch_model-00001-of-00002.bin",
"layers.11.mixer.out_proj.bias": "pytorch_model-00001-of-00002.bin",
"layers.11.mixer.out_proj.weight": "pytorch_model-00001-of-00002.bin",
"layers.11.mlp.fc1.bias": "pytorch_model-00001-of-00002.bin",
"layers.11.mlp.fc1.weight": "pytorch_model-00001-of-00002.bin",
"layers.11.mlp.fc2.bias": "pytorch_model-00001-of-00002.bin",
"layers.11.mlp.fc2.weight": "pytorch_model-00001-of-00002.bin",
"layers.12.ln.bias": "pytorch_model-00001-of-00002.bin",
"layers.12.ln.weight": "pytorch_model-00001-of-00002.bin",
"layers.12.mixer.Wqkv.bias": "pytorch_model-00001-of-00002.bin",
"layers.12.mixer.Wqkv.weight": "pytorch_model-00001-of-00002.bin",
"layers.12.mixer.out_proj.bias": "pytorch_model-00001-of-00002.bin",
"layers.12.mixer.out_proj.weight": "pytorch_model-00001-of-00002.bin",
"layers.12.mlp.fc1.bias": "pytorch_model-00001-of-00002.bin",
"layers.12.mlp.fc1.weight": "pytorch_model-00001-of-00002.bin",
"layers.12.mlp.fc2.bias": "pytorch_model-00001-of-00002.bin",
"layers.12.mlp.fc2.weight": "pytorch_model-00001-of-00002.bin",
"layers.13.ln.bias": "pytorch_model-00001-of-00002.bin",
"layers.13.ln.weight": "pytorch_model-00001-of-00002.bin",
"layers.13.mixer.Wqkv.bias": "pytorch_model-00001-of-00002.bin",
"layers.13.mixer.Wqkv.weight": "pytorch_model-00001-of-00002.bin",
"layers.13.mixer.out_proj.bias": "pytorch_model-00001-of-00002.bin",
"layers.13.mixer.out_proj.weight": "pytorch_model-00001-of-00002.bin",
"layers.13.mlp.fc1.bias": "pytorch_model-00001-of-00002.bin",
"layers.13.mlp.fc1.weight": "pytorch_model-00001-of-00002.bin",
"layers.13.mlp.fc2.bias": "pytorch_model-00001-of-00002.bin",
"layers.13.mlp.fc2.weight": "pytorch_model-00001-of-00002.bin",
"layers.14.ln.bias": "pytorch_model-00001-of-00002.bin",
"layers.14.ln.weight": "pytorch_model-00001-of-00002.bin",
"layers.14.mixer.Wqkv.bias": "pytorch_model-00001-of-00002.bin",
"layers.14.mixer.Wqkv.weight": "pytorch_model-00001-of-00002.bin",
"layers.14.mixer.out_proj.bias": "pytorch_model-00001-of-00002.bin",
"layers.14.mixer.out_proj.weight": "pytorch_model-00001-of-00002.bin",
"layers.14.mlp.fc1.bias": "pytorch_model-00001-of-00002.bin",
"layers.14.mlp.fc1.weight": "pytorch_model-00001-of-00002.bin",
"layers.14.mlp.fc2.bias": "pytorch_model-00001-of-00002.bin",
"layers.14.mlp.fc2.weight": "pytorch_model-00001-of-00002.bin",
"layers.15.ln.bias": "pytorch_model-00001-of-00002.bin",
"layers.15.ln.weight": "pytorch_model-00001-of-00002.bin",
"layers.15.mixer.Wqkv.bias": "pytorch_model-00001-of-00002.bin",
"layers.15.mixer.Wqkv.weight": "pytorch_model-00001-of-00002.bin",
"layers.15.mixer.out_proj.bias": "pytorch_model-00001-of-00002.bin",
"layers.15.mixer.out_proj.weight": "pytorch_model-00001-of-00002.bin",
"layers.15.mlp.fc1.bias": "pytorch_model-00001-of-00002.bin",
"layers.15.mlp.fc1.weight": "pytorch_model-00001-of-00002.bin",
"layers.15.mlp.fc2.bias": "pytorch_model-00001-of-00002.bin",
"layers.15.mlp.fc2.weight": "pytorch_model-00001-of-00002.bin",
"layers.16.ln.bias": "pytorch_model-00001-of-00002.bin",
"layers.16.ln.weight": "pytorch_model-00001-of-00002.bin",
"layers.16.mixer.Wqkv.bias": "pytorch_model-00001-of-00002.bin",
"layers.16.mixer.Wqkv.weight": "pytorch_model-00001-of-00002.bin",
"layers.16.mixer.out_proj.bias": "pytorch_model-00001-of-00002.bin",
"layers.16.mixer.out_proj.weight": "pytorch_model-00001-of-00002.bin",
"layers.16.mlp.fc1.bias": "pytorch_model-00001-of-00002.bin",
"layers.16.mlp.fc1.weight": "pytorch_model-00001-of-00002.bin",
"layers.16.mlp.fc2.bias": "pytorch_model-00001-of-00002.bin",
"layers.16.mlp.fc2.weight": "pytorch_model-00001-of-00002.bin",
"layers.17.ln.bias": "pytorch_model-00001-of-00002.bin",
"layers.17.ln.weight": "pytorch_model-00001-of-00002.bin",
"layers.17.mixer.Wqkv.bias": "pytorch_model-00001-of-00002.bin",
"layers.17.mixer.Wqkv.weight": "pytorch_model-00001-of-00002.bin",
"layers.17.mixer.out_proj.bias": "pytorch_model-00001-of-00002.bin",
"layers.17.mixer.out_proj.weight": "pytorch_model-00001-of-00002.bin",
"layers.17.mlp.fc1.bias": "pytorch_model-00001-of-00002.bin",
"layers.17.mlp.fc1.weight": "pytorch_model-00001-of-00002.bin",
"layers.17.mlp.fc2.bias": "pytorch_model-00001-of-00002.bin",
"layers.17.mlp.fc2.weight": "pytorch_model-00001-of-00002.bin",
"layers.18.ln.bias": "pytorch_model-00001-of-00002.bin",
"layers.18.ln.weight": "pytorch_model-00001-of-00002.bin",
"layers.18.mixer.Wqkv.bias": "pytorch_model-00001-of-00002.bin",
"layers.18.mixer.Wqkv.weight": "pytorch_model-00001-of-00002.bin",
"layers.18.mixer.out_proj.bias": "pytorch_model-00001-of-00002.bin",
"layers.18.mixer.out_proj.weight": "pytorch_model-00001-of-00002.bin",
"layers.18.mlp.fc1.bias": "pytorch_model-00001-of-00002.bin",
"layers.18.mlp.fc1.weight": "pytorch_model-00001-of-00002.bin",
"layers.18.mlp.fc2.bias": "pytorch_model-00001-of-00002.bin",
"layers.18.mlp.fc2.weight": "pytorch_model-00001-of-00002.bin",
"layers.19.ln.bias": "pytorch_model-00001-of-00002.bin",
"layers.19.ln.weight": "pytorch_model-00001-of-00002.bin",
"layers.19.mixer.Wqkv.bias": "pytorch_model-00001-of-00002.bin",
"layers.19.mixer.Wqkv.weight": "pytorch_model-00001-of-00002.bin",
"layers.19.mixer.out_proj.bias": "pytorch_model-00001-of-00002.bin",
"layers.19.mixer.out_proj.weight": "pytorch_model-00001-of-00002.bin",
"layers.19.mlp.fc1.bias": "pytorch_model-00001-of-00002.bin",
"layers.19.mlp.fc1.weight": "pytorch_model-00001-of-00002.bin",
"layers.19.mlp.fc2.bias": "pytorch_model-00001-of-00002.bin",
"layers.19.mlp.fc2.weight": "pytorch_model-00001-of-00002.bin",
"layers.2.ln.bias": "pytorch_model-00001-of-00002.bin",
"layers.2.ln.weight": "pytorch_model-00001-of-00002.bin",
"layers.2.mixer.Wqkv.bias": "pytorch_model-00001-of-00002.bin",
"layers.2.mixer.Wqkv.weight": "pytorch_model-00001-of-00002.bin",
"layers.2.mixer.out_proj.bias": "pytorch_model-00001-of-00002.bin",
"layers.2.mixer.out_proj.weight": "pytorch_model-00001-of-00002.bin",
"layers.2.mlp.fc1.bias": "pytorch_model-00001-of-00002.bin",
"layers.2.mlp.fc1.weight": "pytorch_model-00001-of-00002.bin",
"layers.2.mlp.fc2.bias": "pytorch_model-00001-of-00002.bin",
"layers.2.mlp.fc2.weight": "pytorch_model-00001-of-00002.bin",
"layers.20.ln.bias": "pytorch_model-00001-of-00002.bin",
"layers.20.ln.weight": "pytorch_model-00001-of-00002.bin",
"layers.20.mixer.Wqkv.bias": "pytorch_model-00001-of-00002.bin",
"layers.20.mixer.Wqkv.weight": "pytorch_model-00001-of-00002.bin",
"layers.20.mixer.out_proj.bias": "pytorch_model-00001-of-00002.bin",
"layers.20.mixer.out_proj.weight": "pytorch_model-00001-of-00002.bin",
"layers.20.mlp.fc1.bias": "pytorch_model-00001-of-00002.bin",
"layers.20.mlp.fc1.weight": "pytorch_model-00001-of-00002.bin",
"layers.20.mlp.fc2.bias": "pytorch_model-00001-of-00002.bin",
"layers.20.mlp.fc2.weight": "pytorch_model-00001-of-00002.bin",
"layers.21.ln.bias": "pytorch_model-00001-of-00002.bin",
"layers.21.ln.weight": "pytorch_model-00001-of-00002.bin",
"layers.21.mixer.Wqkv.bias": "pytorch_model-00001-of-00002.bin",
"layers.21.mixer.Wqkv.weight": "pytorch_model-00001-of-00002.bin",
"layers.21.mixer.out_proj.bias": "pytorch_model-00001-of-00002.bin",
"layers.21.mixer.out_proj.weight": "pytorch_model-00001-of-00002.bin",
"layers.21.mlp.fc1.bias": "pytorch_model-00001-of-00002.bin",
"layers.21.mlp.fc1.weight": "pytorch_model-00001-of-00002.bin",
"layers.21.mlp.fc2.bias": "pytorch_model-00001-of-00002.bin",
"layers.21.mlp.fc2.weight": "pytorch_model-00001-of-00002.bin",
"layers.22.ln.bias": "pytorch_model-00001-of-00002.bin",
"layers.22.ln.weight": "pytorch_model-00001-of-00002.bin",
"layers.22.mixer.Wqkv.bias": "pytorch_model-00001-of-00002.bin",
"layers.22.mixer.Wqkv.weight": "pytorch_model-00001-of-00002.bin",
"layers.22.mixer.out_proj.bias": "pytorch_model-00001-of-00002.bin",
"layers.22.mixer.out_proj.weight": "pytorch_model-00001-of-00002.bin",
"layers.22.mlp.fc1.bias": "pytorch_model-00001-of-00002.bin",
"layers.22.mlp.fc1.weight": "pytorch_model-00001-of-00002.bin",
"layers.22.mlp.fc2.bias": "pytorch_model-00001-of-00002.bin",
"layers.22.mlp.fc2.weight": "pytorch_model-00001-of-00002.bin",
"layers.23.ln.bias": "pytorch_model-00001-of-00002.bin",
"layers.23.ln.weight": "pytorch_model-00001-of-00002.bin",
"layers.23.mixer.Wqkv.bias": "pytorch_model-00001-of-00002.bin",
"layers.23.mixer.Wqkv.weight": "pytorch_model-00001-of-00002.bin",
"layers.23.mixer.out_proj.bias": "pytorch_model-00001-of-00002.bin",
"layers.23.mixer.out_proj.weight": "pytorch_model-00001-of-00002.bin",
"layers.23.mlp.fc1.bias": "pytorch_model-00001-of-00002.bin",
"layers.23.mlp.fc1.weight": "pytorch_model-00001-of-00002.bin",
"layers.23.mlp.fc2.bias": "pytorch_model-00001-of-00002.bin",
"layers.23.mlp.fc2.weight": "pytorch_model-00001-of-00002.bin",
"layers.24.ln.bias": "pytorch_model-00001-of-00002.bin",
"layers.24.ln.weight": "pytorch_model-00001-of-00002.bin",
"layers.24.mixer.Wqkv.bias": "pytorch_model-00001-of-00002.bin",
"layers.24.mixer.Wqkv.weight": "pytorch_model-00001-of-00002.bin",
"layers.24.mixer.out_proj.bias": "pytorch_model-00001-of-00002.bin",
"layers.24.mixer.out_proj.weight": "pytorch_model-00001-of-00002.bin",
"layers.24.mlp.fc1.bias": "pytorch_model-00001-of-00002.bin",
"layers.24.mlp.fc1.weight": "pytorch_model-00001-of-00002.bin",
"layers.24.mlp.fc2.bias": "pytorch_model-00001-of-00002.bin",
"layers.24.mlp.fc2.weight": "pytorch_model-00001-of-00002.bin",
"layers.25.ln.bias": "pytorch_model-00001-of-00002.bin",
"layers.25.ln.weight": "pytorch_model-00001-of-00002.bin",
"layers.25.mixer.Wqkv.bias": "pytorch_model-00001-of-00002.bin",
"layers.25.mixer.Wqkv.weight": "pytorch_model-00001-of-00002.bin",
"layers.25.mixer.out_proj.bias": "pytorch_model-00001-of-00002.bin",
"layers.25.mixer.out_proj.weight": "pytorch_model-00001-of-00002.bin",
"layers.25.mlp.fc1.bias": "pytorch_model-00001-of-00002.bin",
"layers.25.mlp.fc1.weight": "pytorch_model-00001-of-00002.bin",
"layers.25.mlp.fc2.bias": "pytorch_model-00001-of-00002.bin",
"layers.25.mlp.fc2.weight": "pytorch_model-00001-of-00002.bin",
"layers.26.ln.bias": "pytorch_model-00001-of-00002.bin",
"layers.26.ln.weight": "pytorch_model-00001-of-00002.bin",
"layers.26.mixer.Wqkv.bias": "pytorch_model-00001-of-00002.bin",
"layers.26.mixer.Wqkv.weight": "pytorch_model-00001-of-00002.bin",
"layers.26.mixer.out_proj.bias": "pytorch_model-00001-of-00002.bin",
"layers.26.mixer.out_proj.weight": "pytorch_model-00001-of-00002.bin",
"layers.26.mlp.fc1.bias": "pytorch_model-00001-of-00002.bin",
"layers.26.mlp.fc1.weight": "pytorch_model-00001-of-00002.bin",
"layers.26.mlp.fc2.bias": "pytorch_model-00001-of-00002.bin",
"layers.26.mlp.fc2.weight": "pytorch_model-00001-of-00002.bin",
"layers.27.ln.bias": "pytorch_model-00001-of-00002.bin",
"layers.27.ln.weight": "pytorch_model-00001-of-00002.bin",
"layers.27.mixer.Wqkv.bias": "pytorch_model-00001-of-00002.bin",
"layers.27.mixer.Wqkv.weight": "pytorch_model-00001-of-00002.bin",
"layers.27.mixer.out_proj.bias": "pytorch_model-00001-of-00002.bin",
"layers.27.mixer.out_proj.weight": "pytorch_model-00001-of-00002.bin",
"layers.27.mlp.fc1.bias": "pytorch_model-00001-of-00002.bin",
"layers.27.mlp.fc1.weight": "pytorch_model-00001-of-00002.bin",
"layers.27.mlp.fc2.bias": "pytorch_model-00001-of-00002.bin",
"layers.27.mlp.fc2.weight": "pytorch_model-00001-of-00002.bin",
"layers.28.ln.bias": "pytorch_model-00001-of-00002.bin",
"layers.28.ln.weight": "pytorch_model-00001-of-00002.bin",
"layers.28.mixer.Wqkv.bias": "pytorch_model-00001-of-00002.bin",
"layers.28.mixer.Wqkv.weight": "pytorch_model-00001-of-00002.bin",
"layers.28.mixer.out_proj.bias": "pytorch_model-00001-of-00002.bin",
"layers.28.mixer.out_proj.weight": "pytorch_model-00001-of-00002.bin",
"layers.28.mlp.fc1.bias": "pytorch_model-00001-of-00002.bin",
"layers.28.mlp.fc1.weight": "pytorch_model-00001-of-00002.bin",
"layers.28.mlp.fc2.bias": "pytorch_model-00001-of-00002.bin",
"layers.28.mlp.fc2.weight": "pytorch_model-00001-of-00002.bin",
"layers.29.ln.bias": "pytorch_model-00001-of-00002.bin",
"layers.29.ln.weight": "pytorch_model-00001-of-00002.bin",
"layers.29.mixer.Wqkv.bias": "pytorch_model-00001-of-00002.bin",
"layers.29.mixer.Wqkv.weight": "pytorch_model-00001-of-00002.bin",
"layers.29.mixer.out_proj.bias": "pytorch_model-00001-of-00002.bin",
"layers.29.mixer.out_proj.weight": "pytorch_model-00001-of-00002.bin",
"layers.29.mlp.fc1.bias": "pytorch_model-00001-of-00002.bin",
"layers.29.mlp.fc1.weight": "pytorch_model-00001-of-00002.bin",
"layers.29.mlp.fc2.bias": "pytorch_model-00001-of-00002.bin",
"layers.29.mlp.fc2.weight": "pytorch_model-00001-of-00002.bin",
"layers.3.ln.bias": "pytorch_model-00001-of-00002.bin",
"layers.3.ln.weight": "pytorch_model-00001-of-00002.bin",
"layers.3.mixer.Wqkv.bias": "pytorch_model-00001-of-00002.bin",
"layers.3.mixer.Wqkv.weight": "pytorch_model-00001-of-00002.bin",
"layers.3.mixer.out_proj.bias": "pytorch_model-00001-of-00002.bin",
"layers.3.mixer.out_proj.weight": "pytorch_model-00001-of-00002.bin",
"layers.3.mlp.fc1.bias": "pytorch_model-00001-of-00002.bin",
"layers.3.mlp.fc1.weight": "pytorch_model-00001-of-00002.bin",
"layers.3.mlp.fc2.bias": "pytorch_model-00001-of-00002.bin",
"layers.3.mlp.fc2.weight": "pytorch_model-00001-of-00002.bin",
"layers.30.ln.bias": "pytorch_model-00001-of-00002.bin",
"layers.30.ln.weight": "pytorch_model-00001-of-00002.bin",
"layers.30.mixer.Wqkv.bias": "pytorch_model-00001-of-00002.bin",
"layers.30.mixer.Wqkv.weight": "pytorch_model-00001-of-00002.bin",
"layers.30.mixer.out_proj.bias": "pytorch_model-00001-of-00002.bin",
"layers.30.mixer.out_proj.weight": "pytorch_model-00001-of-00002.bin",
"layers.30.mlp.fc1.bias": "pytorch_model-00001-of-00002.bin",
"layers.30.mlp.fc1.weight": "pytorch_model-00001-of-00002.bin",
"layers.30.mlp.fc2.bias": "pytorch_model-00001-of-00002.bin",
"layers.30.mlp.fc2.weight": "pytorch_model-00001-of-00002.bin",
"layers.31.ln.bias": "pytorch_model-00001-of-00002.bin",
"layers.31.ln.weight": "pytorch_model-00001-of-00002.bin",
"layers.31.mixer.Wqkv.bias": "pytorch_model-00002-of-00002.bin",
"layers.31.mixer.Wqkv.weight": "pytorch_model-00002-of-00002.bin",
"layers.31.mixer.out_proj.bias": "pytorch_model-00002-of-00002.bin",
"layers.31.mixer.out_proj.weight": "pytorch_model-00002-of-00002.bin",
"layers.31.mlp.fc1.bias": "pytorch_model-00002-of-00002.bin",
"layers.31.mlp.fc1.weight": "pytorch_model-00002-of-00002.bin",
"layers.31.mlp.fc2.bias": "pytorch_model-00002-of-00002.bin",
"layers.31.mlp.fc2.weight": "pytorch_model-00002-of-00002.bin",
"layers.32.ln.bias": "pytorch_model-00002-of-00002.bin",
"layers.32.ln.weight": "pytorch_model-00002-of-00002.bin",
"layers.32.mixer.Wqkv.bias": "pytorch_model-00002-of-00002.bin",
"layers.32.mixer.Wqkv.weight": "pytorch_model-00002-of-00002.bin",
"layers.32.mixer.out_proj.bias": "pytorch_model-00002-of-00002.bin",
"layers.32.mixer.out_proj.weight": "pytorch_model-00002-of-00002.bin",
"layers.32.mlp.fc1.bias": "pytorch_model-00002-of-00002.bin",
"layers.32.mlp.fc1.weight": "pytorch_model-00002-of-00002.bin",
"layers.32.mlp.fc2.bias": "pytorch_model-00002-of-00002.bin",
"layers.32.mlp.fc2.weight": "pytorch_model-00002-of-00002.bin",
"layers.33.linear.bias": "pytorch_model-00002-of-00002.bin",
"layers.33.linear.weight": "pytorch_model-00002-of-00002.bin",
"layers.33.ln.bias": "pytorch_model-00002-of-00002.bin",
"layers.33.ln.weight": "pytorch_model-00002-of-00002.bin",
"layers.4.ln.bias": "pytorch_model-00001-of-00002.bin",
"layers.4.ln.weight": "pytorch_model-00001-of-00002.bin",
"layers.4.mixer.Wqkv.bias": "pytorch_model-00001-of-00002.bin",
"layers.4.mixer.Wqkv.weight": "pytorch_model-00001-of-00002.bin",
"layers.4.mixer.out_proj.bias": "pytorch_model-00001-of-00002.bin",
"layers.4.mixer.out_proj.weight": "pytorch_model-00001-of-00002.bin",
"layers.4.mlp.fc1.bias": "pytorch_model-00001-of-00002.bin",
"layers.4.mlp.fc1.weight": "pytorch_model-00001-of-00002.bin",
"layers.4.mlp.fc2.bias": "pytorch_model-00001-of-00002.bin",
"layers.4.mlp.fc2.weight": "pytorch_model-00001-of-00002.bin",
"layers.5.ln.bias": "pytorch_model-00001-of-00002.bin",
"layers.5.ln.weight": "pytorch_model-00001-of-00002.bin",
"layers.5.mixer.Wqkv.bias": "pytorch_model-00001-of-00002.bin",
"layers.5.mixer.Wqkv.weight": "pytorch_model-00001-of-00002.bin",
"layers.5.mixer.out_proj.bias": "pytorch_model-00001-of-00002.bin",
"layers.5.mixer.out_proj.weight": "pytorch_model-00001-of-00002.bin",
"layers.5.mlp.fc1.bias": "pytorch_model-00001-of-00002.bin",
"layers.5.mlp.fc1.weight": "pytorch_model-00001-of-00002.bin",
"layers.5.mlp.fc2.bias": "pytorch_model-00001-of-00002.bin",
"layers.5.mlp.fc2.weight": "pytorch_model-00001-of-00002.bin",
"layers.6.ln.bias": "pytorch_model-00001-of-00002.bin",
"layers.6.ln.weight": "pytorch_model-00001-of-00002.bin",
"layers.6.mixer.Wqkv.bias": "pytorch_model-00001-of-00002.bin",
"layers.6.mixer.Wqkv.weight": "pytorch_model-00001-of-00002.bin",
"layers.6.mixer.out_proj.bias": "pytorch_model-00001-of-00002.bin",
"layers.6.mixer.out_proj.weight": "pytorch_model-00001-of-00002.bin",
"layers.6.mlp.fc1.bias": "pytorch_model-00001-of-00002.bin",
"layers.6.mlp.fc1.weight": "pytorch_model-00001-of-00002.bin",
"layers.6.mlp.fc2.bias": "pytorch_model-00001-of-00002.bin",
"layers.6.mlp.fc2.weight": "pytorch_model-00001-of-00002.bin",
"layers.7.ln.bias": "pytorch_model-00001-of-00002.bin",
"layers.7.ln.weight": "pytorch_model-00001-of-00002.bin",
"layers.7.mixer.Wqkv.bias": "pytorch_model-00001-of-00002.bin",
"layers.7.mixer.Wqkv.weight": "pytorch_model-00001-of-00002.bin",
"layers.7.mixer.out_proj.bias": "pytorch_model-00001-of-00002.bin",
"layers.7.mixer.out_proj.weight": "pytorch_model-00001-of-00002.bin",
"layers.7.mlp.fc1.bias": "pytorch_model-00001-of-00002.bin",
"layers.7.mlp.fc1.weight": "pytorch_model-00001-of-00002.bin",
"layers.7.mlp.fc2.bias": "pytorch_model-00001-of-00002.bin",
"layers.7.mlp.fc2.weight": "pytorch_model-00001-of-00002.bin",
"layers.8.ln.bias": "pytorch_model-00001-of-00002.bin",
"layers.8.ln.weight": "pytorch_model-00001-of-00002.bin",
"layers.8.mixer.Wqkv.bias": "pytorch_model-00001-of-00002.bin",
"layers.8.mixer.Wqkv.weight": "pytorch_model-00001-of-00002.bin",
"layers.8.mixer.out_proj.bias": "pytorch_model-00001-of-00002.bin",
"layers.8.mixer.out_proj.weight": "pytorch_model-00001-of-00002.bin",
"layers.8.mlp.fc1.bias": "pytorch_model-00001-of-00002.bin",
"layers.8.mlp.fc1.weight": "pytorch_model-00001-of-00002.bin",
"layers.8.mlp.fc2.bias": "pytorch_model-00001-of-00002.bin",
"layers.8.mlp.fc2.weight": "pytorch_model-00001-of-00002.bin",
"layers.9.ln.bias": "pytorch_model-00001-of-00002.bin",
"layers.9.ln.weight": "pytorch_model-00001-of-00002.bin",
"layers.9.mixer.Wqkv.bias": "pytorch_model-00001-of-00002.bin",
"layers.9.mixer.Wqkv.weight": "pytorch_model-00001-of-00002.bin",
"layers.9.mixer.out_proj.bias": "pytorch_model-00001-of-00002.bin",
"layers.9.mixer.out_proj.weight": "pytorch_model-00001-of-00002.bin",
"layers.9.mlp.fc1.bias": "pytorch_model-00001-of-00002.bin",
"layers.9.mlp.fc1.weight": "pytorch_model-00001-of-00002.bin",
"layers.9.mlp.fc2.bias": "pytorch_model-00001-of-00002.bin",
"layers.9.mlp.fc2.weight": "pytorch_model-00001-of-00002.bin"
}
}