SecPaper-Selector-4B / conversion.json
itsalissonsilva's picture
Release evidence selector after all baseline and saved-weight gates passed
78876fc verified
Raw History Blame Contribute Delete
29.2 kB
{
"source_weights_sha256": "532513c9c29daa15fe0040d31db93fb26a82ad92aa4626e6947ed915e0bb878a",
"converted_weights_sha256": "075107999fddb3b2a28f424beac0a329199aa7f6689ed93df3ff14dd674d0a2d",
"base_revision": "851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a",
"source_tensors": 304,
"converted_tensors": 208,
"qkv_groups": 24,
"qkv_rank": 48,
"scaling_preserved": true,
"all_source_tensors_consumed": true,
"max_float32_update_error": 1.862645149230957e-09,
"conversion": "A concatenation, B block diagonal, rank/alpha tripled; output projection -> lm_head",
"mapping": {
"base_model.model.model.layers.0.linear_attn": "base_model.model.model.language_model.layers.0.linear_attn.in_proj_qkv",
"base_model.model.model.layers.1.linear_attn": "base_model.model.model.language_model.layers.1.linear_attn.in_proj_qkv",
"base_model.model.model.layers.10.linear_attn": "base_model.model.model.language_model.layers.10.linear_attn.in_proj_qkv",
"base_model.model.model.layers.12.linear_attn": "base_model.model.model.language_model.layers.12.linear_attn.in_proj_qkv",
"base_model.model.model.layers.13.linear_attn": "base_model.model.model.language_model.layers.13.linear_attn.in_proj_qkv",
"base_model.model.model.layers.14.linear_attn": "base_model.model.model.language_model.layers.14.linear_attn.in_proj_qkv",
"base_model.model.model.layers.16.linear_attn": "base_model.model.model.language_model.layers.16.linear_attn.in_proj_qkv",
"base_model.model.model.layers.17.linear_attn": "base_model.model.model.language_model.layers.17.linear_attn.in_proj_qkv",
"base_model.model.model.layers.18.linear_attn": "base_model.model.model.language_model.layers.18.linear_attn.in_proj_qkv",
"base_model.model.model.layers.2.linear_attn": "base_model.model.model.language_model.layers.2.linear_attn.in_proj_qkv",
"base_model.model.model.layers.20.linear_attn": "base_model.model.model.language_model.layers.20.linear_attn.in_proj_qkv",
"base_model.model.model.layers.21.linear_attn": "base_model.model.model.language_model.layers.21.linear_attn.in_proj_qkv",
"base_model.model.model.layers.22.linear_attn": "base_model.model.model.language_model.layers.22.linear_attn.in_proj_qkv",
"base_model.model.model.layers.24.linear_attn": "base_model.model.model.language_model.layers.24.linear_attn.in_proj_qkv",
"base_model.model.model.layers.25.linear_attn": "base_model.model.model.language_model.layers.25.linear_attn.in_proj_qkv",
"base_model.model.model.layers.26.linear_attn": "base_model.model.model.language_model.layers.26.linear_attn.in_proj_qkv",
"base_model.model.model.layers.28.linear_attn": "base_model.model.model.language_model.layers.28.linear_attn.in_proj_qkv",
"base_model.model.model.layers.29.linear_attn": "base_model.model.model.language_model.layers.29.linear_attn.in_proj_qkv",
"base_model.model.model.layers.30.linear_attn": "base_model.model.model.language_model.layers.30.linear_attn.in_proj_qkv",
"base_model.model.model.layers.4.linear_attn": "base_model.model.model.language_model.layers.4.linear_attn.in_proj_qkv",
"base_model.model.model.layers.5.linear_attn": "base_model.model.model.language_model.layers.5.linear_attn.in_proj_qkv",
"base_model.model.model.layers.6.linear_attn": "base_model.model.model.language_model.layers.6.linear_attn.in_proj_qkv",
"base_model.model.model.layers.8.linear_attn": "base_model.model.model.language_model.layers.8.linear_attn.in_proj_qkv",
"base_model.model.model.layers.9.linear_attn": "base_model.model.model.language_model.layers.9.linear_attn.in_proj_qkv",
"base_model.model.model.layers.0.linear_attn.in_proj_z.lora_A.weight": "base_model.model.model.language_model.layers.0.linear_attn.in_proj_z.lora_A.weight",
"base_model.model.model.layers.0.linear_attn.in_proj_z.lora_B.weight": "base_model.model.model.language_model.layers.0.linear_attn.in_proj_z.lora_B.weight",
"base_model.model.model.layers.0.linear_attn.out_proj.lora_A.weight": "base_model.model.model.language_model.layers.0.linear_attn.out_proj.lora_A.weight",
"base_model.model.model.layers.0.linear_attn.out_proj.lora_B.weight": "base_model.model.model.language_model.layers.0.linear_attn.out_proj.lora_B.weight",
"base_model.model.model.layers.1.linear_attn.in_proj_z.lora_A.weight": "base_model.model.model.language_model.layers.1.linear_attn.in_proj_z.lora_A.weight",
"base_model.model.model.layers.1.linear_attn.in_proj_z.lora_B.weight": "base_model.model.model.language_model.layers.1.linear_attn.in_proj_z.lora_B.weight",
"base_model.model.model.layers.1.linear_attn.out_proj.lora_A.weight": "base_model.model.model.language_model.layers.1.linear_attn.out_proj.lora_A.weight",
"base_model.model.model.layers.1.linear_attn.out_proj.lora_B.weight": "base_model.model.model.language_model.layers.1.linear_attn.out_proj.lora_B.weight",
"base_model.model.model.layers.2.linear_attn.in_proj_z.lora_A.weight": "base_model.model.model.language_model.layers.2.linear_attn.in_proj_z.lora_A.weight",
"base_model.model.model.layers.2.linear_attn.in_proj_z.lora_B.weight": "base_model.model.model.language_model.layers.2.linear_attn.in_proj_z.lora_B.weight",
"base_model.model.model.layers.2.linear_attn.out_proj.lora_A.weight": "base_model.model.model.language_model.layers.2.linear_attn.out_proj.lora_A.weight",
"base_model.model.model.layers.2.linear_attn.out_proj.lora_B.weight": "base_model.model.model.language_model.layers.2.linear_attn.out_proj.lora_B.weight",
"base_model.model.model.layers.3.self_attn.q_proj.lora_A.weight": "base_model.model.model.language_model.layers.3.self_attn.q_proj.lora_A.weight",
"base_model.model.model.layers.3.self_attn.q_proj.lora_B.weight": "base_model.model.model.language_model.layers.3.self_attn.q_proj.lora_B.weight",
"base_model.model.model.layers.3.self_attn.k_proj.lora_A.weight": "base_model.model.model.language_model.layers.3.self_attn.k_proj.lora_A.weight",
"base_model.model.model.layers.3.self_attn.k_proj.lora_B.weight": "base_model.model.model.language_model.layers.3.self_attn.k_proj.lora_B.weight",
"base_model.model.model.layers.3.self_attn.v_proj.lora_A.weight": "base_model.model.model.language_model.layers.3.self_attn.v_proj.lora_A.weight",
"base_model.model.model.layers.3.self_attn.v_proj.lora_B.weight": "base_model.model.model.language_model.layers.3.self_attn.v_proj.lora_B.weight",
"base_model.model.model.layers.3.self_attn.o_proj.lora_A.weight": "base_model.model.model.language_model.layers.3.self_attn.o_proj.lora_A.weight",
"base_model.model.model.layers.3.self_attn.o_proj.lora_B.weight": "base_model.model.model.language_model.layers.3.self_attn.o_proj.lora_B.weight",
"base_model.model.model.layers.4.linear_attn.in_proj_z.lora_A.weight": "base_model.model.model.language_model.layers.4.linear_attn.in_proj_z.lora_A.weight",
"base_model.model.model.layers.4.linear_attn.in_proj_z.lora_B.weight": "base_model.model.model.language_model.layers.4.linear_attn.in_proj_z.lora_B.weight",
"base_model.model.model.layers.4.linear_attn.out_proj.lora_A.weight": "base_model.model.model.language_model.layers.4.linear_attn.out_proj.lora_A.weight",
"base_model.model.model.layers.4.linear_attn.out_proj.lora_B.weight": "base_model.model.model.language_model.layers.4.linear_attn.out_proj.lora_B.weight",
"base_model.model.model.layers.5.linear_attn.in_proj_z.lora_A.weight": "base_model.model.model.language_model.layers.5.linear_attn.in_proj_z.lora_A.weight",
"base_model.model.model.layers.5.linear_attn.in_proj_z.lora_B.weight": "base_model.model.model.language_model.layers.5.linear_attn.in_proj_z.lora_B.weight",
"base_model.model.model.layers.5.linear_attn.out_proj.lora_A.weight": "base_model.model.model.language_model.layers.5.linear_attn.out_proj.lora_A.weight",
"base_model.model.model.layers.5.linear_attn.out_proj.lora_B.weight": "base_model.model.model.language_model.layers.5.linear_attn.out_proj.lora_B.weight",
"base_model.model.model.layers.6.linear_attn.in_proj_z.lora_A.weight": "base_model.model.model.language_model.layers.6.linear_attn.in_proj_z.lora_A.weight",
"base_model.model.model.layers.6.linear_attn.in_proj_z.lora_B.weight": "base_model.model.model.language_model.layers.6.linear_attn.in_proj_z.lora_B.weight",
"base_model.model.model.layers.6.linear_attn.out_proj.lora_A.weight": "base_model.model.model.language_model.layers.6.linear_attn.out_proj.lora_A.weight",
"base_model.model.model.layers.6.linear_attn.out_proj.lora_B.weight": "base_model.model.model.language_model.layers.6.linear_attn.out_proj.lora_B.weight",
"base_model.model.model.layers.7.self_attn.q_proj.lora_A.weight": "base_model.model.model.language_model.layers.7.self_attn.q_proj.lora_A.weight",
"base_model.model.model.layers.7.self_attn.q_proj.lora_B.weight": "base_model.model.model.language_model.layers.7.self_attn.q_proj.lora_B.weight",
"base_model.model.model.layers.7.self_attn.k_proj.lora_A.weight": "base_model.model.model.language_model.layers.7.self_attn.k_proj.lora_A.weight",
"base_model.model.model.layers.7.self_attn.k_proj.lora_B.weight": "base_model.model.model.language_model.layers.7.self_attn.k_proj.lora_B.weight",
"base_model.model.model.layers.7.self_attn.v_proj.lora_A.weight": "base_model.model.model.language_model.layers.7.self_attn.v_proj.lora_A.weight",
"base_model.model.model.layers.7.self_attn.v_proj.lora_B.weight": "base_model.model.model.language_model.layers.7.self_attn.v_proj.lora_B.weight",
"base_model.model.model.layers.7.self_attn.o_proj.lora_A.weight": "base_model.model.model.language_model.layers.7.self_attn.o_proj.lora_A.weight",
"base_model.model.model.layers.7.self_attn.o_proj.lora_B.weight": "base_model.model.model.language_model.layers.7.self_attn.o_proj.lora_B.weight",
"base_model.model.model.layers.8.linear_attn.in_proj_z.lora_A.weight": "base_model.model.model.language_model.layers.8.linear_attn.in_proj_z.lora_A.weight",
"base_model.model.model.layers.8.linear_attn.in_proj_z.lora_B.weight": "base_model.model.model.language_model.layers.8.linear_attn.in_proj_z.lora_B.weight",
"base_model.model.model.layers.8.linear_attn.out_proj.lora_A.weight": "base_model.model.model.language_model.layers.8.linear_attn.out_proj.lora_A.weight",
"base_model.model.model.layers.8.linear_attn.out_proj.lora_B.weight": "base_model.model.model.language_model.layers.8.linear_attn.out_proj.lora_B.weight",
"base_model.model.model.layers.9.linear_attn.in_proj_z.lora_A.weight": "base_model.model.model.language_model.layers.9.linear_attn.in_proj_z.lora_A.weight",
"base_model.model.model.layers.9.linear_attn.in_proj_z.lora_B.weight": "base_model.model.model.language_model.layers.9.linear_attn.in_proj_z.lora_B.weight",
"base_model.model.model.layers.9.linear_attn.out_proj.lora_A.weight": "base_model.model.model.language_model.layers.9.linear_attn.out_proj.lora_A.weight",
"base_model.model.model.layers.9.linear_attn.out_proj.lora_B.weight": "base_model.model.model.language_model.layers.9.linear_attn.out_proj.lora_B.weight",
"base_model.model.model.layers.10.linear_attn.in_proj_z.lora_A.weight": "base_model.model.model.language_model.layers.10.linear_attn.in_proj_z.lora_A.weight",
"base_model.model.model.layers.10.linear_attn.in_proj_z.lora_B.weight": "base_model.model.model.language_model.layers.10.linear_attn.in_proj_z.lora_B.weight",
"base_model.model.model.layers.10.linear_attn.out_proj.lora_A.weight": "base_model.model.model.language_model.layers.10.linear_attn.out_proj.lora_A.weight",
"base_model.model.model.layers.10.linear_attn.out_proj.lora_B.weight": "base_model.model.model.language_model.layers.10.linear_attn.out_proj.lora_B.weight",
"base_model.model.model.layers.11.self_attn.q_proj.lora_A.weight": "base_model.model.model.language_model.layers.11.self_attn.q_proj.lora_A.weight",
"base_model.model.model.layers.11.self_attn.q_proj.lora_B.weight": "base_model.model.model.language_model.layers.11.self_attn.q_proj.lora_B.weight",
"base_model.model.model.layers.11.self_attn.k_proj.lora_A.weight": "base_model.model.model.language_model.layers.11.self_attn.k_proj.lora_A.weight",
"base_model.model.model.layers.11.self_attn.k_proj.lora_B.weight": "base_model.model.model.language_model.layers.11.self_attn.k_proj.lora_B.weight",
"base_model.model.model.layers.11.self_attn.v_proj.lora_A.weight": "base_model.model.model.language_model.layers.11.self_attn.v_proj.lora_A.weight",
"base_model.model.model.layers.11.self_attn.v_proj.lora_B.weight": "base_model.model.model.language_model.layers.11.self_attn.v_proj.lora_B.weight",
"base_model.model.model.layers.11.self_attn.o_proj.lora_A.weight": "base_model.model.model.language_model.layers.11.self_attn.o_proj.lora_A.weight",
"base_model.model.model.layers.11.self_attn.o_proj.lora_B.weight": "base_model.model.model.language_model.layers.11.self_attn.o_proj.lora_B.weight",
"base_model.model.model.layers.12.linear_attn.in_proj_z.lora_A.weight": "base_model.model.model.language_model.layers.12.linear_attn.in_proj_z.lora_A.weight",
"base_model.model.model.layers.12.linear_attn.in_proj_z.lora_B.weight": "base_model.model.model.language_model.layers.12.linear_attn.in_proj_z.lora_B.weight",
"base_model.model.model.layers.12.linear_attn.out_proj.lora_A.weight": "base_model.model.model.language_model.layers.12.linear_attn.out_proj.lora_A.weight",
"base_model.model.model.layers.12.linear_attn.out_proj.lora_B.weight": "base_model.model.model.language_model.layers.12.linear_attn.out_proj.lora_B.weight",
"base_model.model.model.layers.13.linear_attn.in_proj_z.lora_A.weight": "base_model.model.model.language_model.layers.13.linear_attn.in_proj_z.lora_A.weight",
"base_model.model.model.layers.13.linear_attn.in_proj_z.lora_B.weight": "base_model.model.model.language_model.layers.13.linear_attn.in_proj_z.lora_B.weight",
"base_model.model.model.layers.13.linear_attn.out_proj.lora_A.weight": "base_model.model.model.language_model.layers.13.linear_attn.out_proj.lora_A.weight",
"base_model.model.model.layers.13.linear_attn.out_proj.lora_B.weight": "base_model.model.model.language_model.layers.13.linear_attn.out_proj.lora_B.weight",
"base_model.model.model.layers.14.linear_attn.in_proj_z.lora_A.weight": "base_model.model.model.language_model.layers.14.linear_attn.in_proj_z.lora_A.weight",
"base_model.model.model.layers.14.linear_attn.in_proj_z.lora_B.weight": "base_model.model.model.language_model.layers.14.linear_attn.in_proj_z.lora_B.weight",
"base_model.model.model.layers.14.linear_attn.out_proj.lora_A.weight": "base_model.model.model.language_model.layers.14.linear_attn.out_proj.lora_A.weight",
"base_model.model.model.layers.14.linear_attn.out_proj.lora_B.weight": "base_model.model.model.language_model.layers.14.linear_attn.out_proj.lora_B.weight",
"base_model.model.model.layers.15.self_attn.q_proj.lora_A.weight": "base_model.model.model.language_model.layers.15.self_attn.q_proj.lora_A.weight",
"base_model.model.model.layers.15.self_attn.q_proj.lora_B.weight": "base_model.model.model.language_model.layers.15.self_attn.q_proj.lora_B.weight",
"base_model.model.model.layers.15.self_attn.k_proj.lora_A.weight": "base_model.model.model.language_model.layers.15.self_attn.k_proj.lora_A.weight",
"base_model.model.model.layers.15.self_attn.k_proj.lora_B.weight": "base_model.model.model.language_model.layers.15.self_attn.k_proj.lora_B.weight",
"base_model.model.model.layers.15.self_attn.v_proj.lora_A.weight": "base_model.model.model.language_model.layers.15.self_attn.v_proj.lora_A.weight",
"base_model.model.model.layers.15.self_attn.v_proj.lora_B.weight": "base_model.model.model.language_model.layers.15.self_attn.v_proj.lora_B.weight",
"base_model.model.model.layers.15.self_attn.o_proj.lora_A.weight": "base_model.model.model.language_model.layers.15.self_attn.o_proj.lora_A.weight",
"base_model.model.model.layers.15.self_attn.o_proj.lora_B.weight": "base_model.model.model.language_model.layers.15.self_attn.o_proj.lora_B.weight",
"base_model.model.model.layers.16.linear_attn.in_proj_z.lora_A.weight": "base_model.model.model.language_model.layers.16.linear_attn.in_proj_z.lora_A.weight",
"base_model.model.model.layers.16.linear_attn.in_proj_z.lora_B.weight": "base_model.model.model.language_model.layers.16.linear_attn.in_proj_z.lora_B.weight",
"base_model.model.model.layers.16.linear_attn.out_proj.lora_A.weight": "base_model.model.model.language_model.layers.16.linear_attn.out_proj.lora_A.weight",
"base_model.model.model.layers.16.linear_attn.out_proj.lora_B.weight": "base_model.model.model.language_model.layers.16.linear_attn.out_proj.lora_B.weight",
"base_model.model.model.layers.17.linear_attn.in_proj_z.lora_A.weight": "base_model.model.model.language_model.layers.17.linear_attn.in_proj_z.lora_A.weight",
"base_model.model.model.layers.17.linear_attn.in_proj_z.lora_B.weight": "base_model.model.model.language_model.layers.17.linear_attn.in_proj_z.lora_B.weight",
"base_model.model.model.layers.17.linear_attn.out_proj.lora_A.weight": "base_model.model.model.language_model.layers.17.linear_attn.out_proj.lora_A.weight",
"base_model.model.model.layers.17.linear_attn.out_proj.lora_B.weight": "base_model.model.model.language_model.layers.17.linear_attn.out_proj.lora_B.weight",
"base_model.model.model.layers.18.linear_attn.in_proj_z.lora_A.weight": "base_model.model.model.language_model.layers.18.linear_attn.in_proj_z.lora_A.weight",
"base_model.model.model.layers.18.linear_attn.in_proj_z.lora_B.weight": "base_model.model.model.language_model.layers.18.linear_attn.in_proj_z.lora_B.weight",
"base_model.model.model.layers.18.linear_attn.out_proj.lora_A.weight": "base_model.model.model.language_model.layers.18.linear_attn.out_proj.lora_A.weight",
"base_model.model.model.layers.18.linear_attn.out_proj.lora_B.weight": "base_model.model.model.language_model.layers.18.linear_attn.out_proj.lora_B.weight",
"base_model.model.model.layers.19.self_attn.q_proj.lora_A.weight": "base_model.model.model.language_model.layers.19.self_attn.q_proj.lora_A.weight",
"base_model.model.model.layers.19.self_attn.q_proj.lora_B.weight": "base_model.model.model.language_model.layers.19.self_attn.q_proj.lora_B.weight",
"base_model.model.model.layers.19.self_attn.k_proj.lora_A.weight": "base_model.model.model.language_model.layers.19.self_attn.k_proj.lora_A.weight",
"base_model.model.model.layers.19.self_attn.k_proj.lora_B.weight": "base_model.model.model.language_model.layers.19.self_attn.k_proj.lora_B.weight",
"base_model.model.model.layers.19.self_attn.v_proj.lora_A.weight": "base_model.model.model.language_model.layers.19.self_attn.v_proj.lora_A.weight",
"base_model.model.model.layers.19.self_attn.v_proj.lora_B.weight": "base_model.model.model.language_model.layers.19.self_attn.v_proj.lora_B.weight",
"base_model.model.model.layers.19.self_attn.o_proj.lora_A.weight": "base_model.model.model.language_model.layers.19.self_attn.o_proj.lora_A.weight",
"base_model.model.model.layers.19.self_attn.o_proj.lora_B.weight": "base_model.model.model.language_model.layers.19.self_attn.o_proj.lora_B.weight",
"base_model.model.model.layers.20.linear_attn.in_proj_z.lora_A.weight": "base_model.model.model.language_model.layers.20.linear_attn.in_proj_z.lora_A.weight",
"base_model.model.model.layers.20.linear_attn.in_proj_z.lora_B.weight": "base_model.model.model.language_model.layers.20.linear_attn.in_proj_z.lora_B.weight",
"base_model.model.model.layers.20.linear_attn.out_proj.lora_A.weight": "base_model.model.model.language_model.layers.20.linear_attn.out_proj.lora_A.weight",
"base_model.model.model.layers.20.linear_attn.out_proj.lora_B.weight": "base_model.model.model.language_model.layers.20.linear_attn.out_proj.lora_B.weight",
"base_model.model.model.layers.21.linear_attn.in_proj_z.lora_A.weight": "base_model.model.model.language_model.layers.21.linear_attn.in_proj_z.lora_A.weight",
"base_model.model.model.layers.21.linear_attn.in_proj_z.lora_B.weight": "base_model.model.model.language_model.layers.21.linear_attn.in_proj_z.lora_B.weight",
"base_model.model.model.layers.21.linear_attn.out_proj.lora_A.weight": "base_model.model.model.language_model.layers.21.linear_attn.out_proj.lora_A.weight",
"base_model.model.model.layers.21.linear_attn.out_proj.lora_B.weight": "base_model.model.model.language_model.layers.21.linear_attn.out_proj.lora_B.weight",
"base_model.model.model.layers.22.linear_attn.in_proj_z.lora_A.weight": "base_model.model.model.language_model.layers.22.linear_attn.in_proj_z.lora_A.weight",
"base_model.model.model.layers.22.linear_attn.in_proj_z.lora_B.weight": "base_model.model.model.language_model.layers.22.linear_attn.in_proj_z.lora_B.weight",
"base_model.model.model.layers.22.linear_attn.out_proj.lora_A.weight": "base_model.model.model.language_model.layers.22.linear_attn.out_proj.lora_A.weight",
"base_model.model.model.layers.22.linear_attn.out_proj.lora_B.weight": "base_model.model.model.language_model.layers.22.linear_attn.out_proj.lora_B.weight",
"base_model.model.model.layers.23.self_attn.q_proj.lora_A.weight": "base_model.model.model.language_model.layers.23.self_attn.q_proj.lora_A.weight",
"base_model.model.model.layers.23.self_attn.q_proj.lora_B.weight": "base_model.model.model.language_model.layers.23.self_attn.q_proj.lora_B.weight",
"base_model.model.model.layers.23.self_attn.k_proj.lora_A.weight": "base_model.model.model.language_model.layers.23.self_attn.k_proj.lora_A.weight",
"base_model.model.model.layers.23.self_attn.k_proj.lora_B.weight": "base_model.model.model.language_model.layers.23.self_attn.k_proj.lora_B.weight",
"base_model.model.model.layers.23.self_attn.v_proj.lora_A.weight": "base_model.model.model.language_model.layers.23.self_attn.v_proj.lora_A.weight",
"base_model.model.model.layers.23.self_attn.v_proj.lora_B.weight": "base_model.model.model.language_model.layers.23.self_attn.v_proj.lora_B.weight",
"base_model.model.model.layers.23.self_attn.o_proj.lora_A.weight": "base_model.model.model.language_model.layers.23.self_attn.o_proj.lora_A.weight",
"base_model.model.model.layers.23.self_attn.o_proj.lora_B.weight": "base_model.model.model.language_model.layers.23.self_attn.o_proj.lora_B.weight",
"base_model.model.model.layers.24.linear_attn.in_proj_z.lora_A.weight": "base_model.model.model.language_model.layers.24.linear_attn.in_proj_z.lora_A.weight",
"base_model.model.model.layers.24.linear_attn.in_proj_z.lora_B.weight": "base_model.model.model.language_model.layers.24.linear_attn.in_proj_z.lora_B.weight",
"base_model.model.model.layers.24.linear_attn.out_proj.lora_A.weight": "base_model.model.model.language_model.layers.24.linear_attn.out_proj.lora_A.weight",
"base_model.model.model.layers.24.linear_attn.out_proj.lora_B.weight": "base_model.model.model.language_model.layers.24.linear_attn.out_proj.lora_B.weight",
"base_model.model.model.layers.25.linear_attn.in_proj_z.lora_A.weight": "base_model.model.model.language_model.layers.25.linear_attn.in_proj_z.lora_A.weight",
"base_model.model.model.layers.25.linear_attn.in_proj_z.lora_B.weight": "base_model.model.model.language_model.layers.25.linear_attn.in_proj_z.lora_B.weight",
"base_model.model.model.layers.25.linear_attn.out_proj.lora_A.weight": "base_model.model.model.language_model.layers.25.linear_attn.out_proj.lora_A.weight",
"base_model.model.model.layers.25.linear_attn.out_proj.lora_B.weight": "base_model.model.model.language_model.layers.25.linear_attn.out_proj.lora_B.weight",
"base_model.model.model.layers.26.linear_attn.in_proj_z.lora_A.weight": "base_model.model.model.language_model.layers.26.linear_attn.in_proj_z.lora_A.weight",
"base_model.model.model.layers.26.linear_attn.in_proj_z.lora_B.weight": "base_model.model.model.language_model.layers.26.linear_attn.in_proj_z.lora_B.weight",
"base_model.model.model.layers.26.linear_attn.out_proj.lora_A.weight": "base_model.model.model.language_model.layers.26.linear_attn.out_proj.lora_A.weight",
"base_model.model.model.layers.26.linear_attn.out_proj.lora_B.weight": "base_model.model.model.language_model.layers.26.linear_attn.out_proj.lora_B.weight",
"base_model.model.model.layers.27.self_attn.q_proj.lora_A.weight": "base_model.model.model.language_model.layers.27.self_attn.q_proj.lora_A.weight",
"base_model.model.model.layers.27.self_attn.q_proj.lora_B.weight": "base_model.model.model.language_model.layers.27.self_attn.q_proj.lora_B.weight",
"base_model.model.model.layers.27.self_attn.k_proj.lora_A.weight": "base_model.model.model.language_model.layers.27.self_attn.k_proj.lora_A.weight",
"base_model.model.model.layers.27.self_attn.k_proj.lora_B.weight": "base_model.model.model.language_model.layers.27.self_attn.k_proj.lora_B.weight",
"base_model.model.model.layers.27.self_attn.v_proj.lora_A.weight": "base_model.model.model.language_model.layers.27.self_attn.v_proj.lora_A.weight",
"base_model.model.model.layers.27.self_attn.v_proj.lora_B.weight": "base_model.model.model.language_model.layers.27.self_attn.v_proj.lora_B.weight",
"base_model.model.model.layers.27.self_attn.o_proj.lora_A.weight": "base_model.model.model.language_model.layers.27.self_attn.o_proj.lora_A.weight",
"base_model.model.model.layers.27.self_attn.o_proj.lora_B.weight": "base_model.model.model.language_model.layers.27.self_attn.o_proj.lora_B.weight",
"base_model.model.model.layers.28.linear_attn.in_proj_z.lora_A.weight": "base_model.model.model.language_model.layers.28.linear_attn.in_proj_z.lora_A.weight",
"base_model.model.model.layers.28.linear_attn.in_proj_z.lora_B.weight": "base_model.model.model.language_model.layers.28.linear_attn.in_proj_z.lora_B.weight",
"base_model.model.model.layers.28.linear_attn.out_proj.lora_A.weight": "base_model.model.model.language_model.layers.28.linear_attn.out_proj.lora_A.weight",
"base_model.model.model.layers.28.linear_attn.out_proj.lora_B.weight": "base_model.model.model.language_model.layers.28.linear_attn.out_proj.lora_B.weight",
"base_model.model.model.layers.29.linear_attn.in_proj_z.lora_A.weight": "base_model.model.model.language_model.layers.29.linear_attn.in_proj_z.lora_A.weight",
"base_model.model.model.layers.29.linear_attn.in_proj_z.lora_B.weight": "base_model.model.model.language_model.layers.29.linear_attn.in_proj_z.lora_B.weight",
"base_model.model.model.layers.29.linear_attn.out_proj.lora_A.weight": "base_model.model.model.language_model.layers.29.linear_attn.out_proj.lora_A.weight",
"base_model.model.model.layers.29.linear_attn.out_proj.lora_B.weight": "base_model.model.model.language_model.layers.29.linear_attn.out_proj.lora_B.weight",
"base_model.model.model.layers.30.linear_attn.in_proj_z.lora_A.weight": "base_model.model.model.language_model.layers.30.linear_attn.in_proj_z.lora_A.weight",
"base_model.model.model.layers.30.linear_attn.in_proj_z.lora_B.weight": "base_model.model.model.language_model.layers.30.linear_attn.in_proj_z.lora_B.weight",
"base_model.model.model.layers.30.linear_attn.out_proj.lora_A.weight": "base_model.model.model.language_model.layers.30.linear_attn.out_proj.lora_A.weight",
"base_model.model.model.layers.30.linear_attn.out_proj.lora_B.weight": "base_model.model.model.language_model.layers.30.linear_attn.out_proj.lora_B.weight",
"base_model.model.model.layers.31.self_attn.q_proj.lora_A.weight": "base_model.model.model.language_model.layers.31.self_attn.q_proj.lora_A.weight",
"base_model.model.model.layers.31.self_attn.q_proj.lora_B.weight": "base_model.model.model.language_model.layers.31.self_attn.q_proj.lora_B.weight",
"base_model.model.model.layers.31.self_attn.k_proj.lora_A.weight": "base_model.model.model.language_model.layers.31.self_attn.k_proj.lora_A.weight",
"base_model.model.model.layers.31.self_attn.k_proj.lora_B.weight": "base_model.model.model.language_model.layers.31.self_attn.k_proj.lora_B.weight",
"base_model.model.model.layers.31.self_attn.v_proj.lora_A.weight": "base_model.model.model.language_model.layers.31.self_attn.v_proj.lora_A.weight",
"base_model.model.model.layers.31.self_attn.v_proj.lora_B.weight": "base_model.model.model.language_model.layers.31.self_attn.v_proj.lora_B.weight",
"base_model.model.model.layers.31.self_attn.o_proj.lora_A.weight": "base_model.model.model.language_model.layers.31.self_attn.o_proj.lora_A.weight",
"base_model.model.model.layers.31.self_attn.o_proj.lora_B.weight": "base_model.model.model.language_model.layers.31.self_attn.o_proj.lora_B.weight"
},
"external_load_verified": true,
"verification_report": "evaluation.json"
}