{ "artifact_format": "orbitquant-v1", "source_model_id": "Wan-AI/Wan2.1-T2V-1.3B-Diffusers", "source_revision": "0fad780a534b6463e45facd96134c9f345acfa5b", "source_license": "apache-2.0", "quant_method": "orbitquant", "paper": "https://arxiv.org/abs/2607.02461", "weight_bits": 4, "activation_bits": 4, "rotation": "rpbh", "rotation_seed": 0, "block_size": "paper", "block_size_policy": "largest_power_of_two_dividing_dim", "codebook": "lloyd_max", "codebook_version": 2, "row_norm_dtype": "bfloat16", "runtime_mode": "auto_fused", "activation_kernel_backend": "auto", "activation_eps": 1e-10, "adaln_group_size": 64, "quantization_device": "cuda", "weight_quantization_backend": "triton_cuda", "quantization_staging_mode": "streaming", "target_policy": "wan", "adaln_policy": "int4_rtn_group64_bf16_activation", "quantized_modules": [ "blocks.0.attn1.to_q", "blocks.0.attn1.to_k", "blocks.0.attn1.to_v", "blocks.0.attn1.to_out.0", "blocks.0.attn2.to_q", "blocks.0.attn2.to_k", "blocks.0.attn2.to_v", "blocks.0.attn2.to_out.0", "blocks.0.ffn.net.0.proj", "blocks.0.ffn.net.2", "blocks.1.attn1.to_q", "blocks.1.attn1.to_k", "blocks.1.attn1.to_v", "blocks.1.attn1.to_out.0", "blocks.1.attn2.to_q", "blocks.1.attn2.to_k", "blocks.1.attn2.to_v", "blocks.1.attn2.to_out.0", "blocks.1.ffn.net.0.proj", "blocks.1.ffn.net.2", "blocks.2.attn1.to_q", "blocks.2.attn1.to_k", "blocks.2.attn1.to_v", "blocks.2.attn1.to_out.0", "blocks.2.attn2.to_q", "blocks.2.attn2.to_k", "blocks.2.attn2.to_v", "blocks.2.attn2.to_out.0", "blocks.2.ffn.net.0.proj", "blocks.2.ffn.net.2", "blocks.3.attn1.to_q", "blocks.3.attn1.to_k", "blocks.3.attn1.to_v", "blocks.3.attn1.to_out.0", "blocks.3.attn2.to_q", "blocks.3.attn2.to_k", "blocks.3.attn2.to_v", "blocks.3.attn2.to_out.0", "blocks.3.ffn.net.0.proj", "blocks.3.ffn.net.2", "blocks.4.attn1.to_q", "blocks.4.attn1.to_k", "blocks.4.attn1.to_v", "blocks.4.attn1.to_out.0", "blocks.4.attn2.to_q", "blocks.4.attn2.to_k", "blocks.4.attn2.to_v", "blocks.4.attn2.to_out.0", "blocks.4.ffn.net.0.proj", "blocks.4.ffn.net.2", "blocks.5.attn1.to_q", "blocks.5.attn1.to_k", "blocks.5.attn1.to_v", "blocks.5.attn1.to_out.0", "blocks.5.attn2.to_q", "blocks.5.attn2.to_k", "blocks.5.attn2.to_v", "blocks.5.attn2.to_out.0", "blocks.5.ffn.net.0.proj", "blocks.5.ffn.net.2", "blocks.6.attn1.to_q", "blocks.6.attn1.to_k", "blocks.6.attn1.to_v", "blocks.6.attn1.to_out.0", "blocks.6.attn2.to_q", "blocks.6.attn2.to_k", "blocks.6.attn2.to_v", "blocks.6.attn2.to_out.0", "blocks.6.ffn.net.0.proj", "blocks.6.ffn.net.2", "blocks.7.attn1.to_q", "blocks.7.attn1.to_k", "blocks.7.attn1.to_v", "blocks.7.attn1.to_out.0", "blocks.7.attn2.to_q", "blocks.7.attn2.to_k", "blocks.7.attn2.to_v", "blocks.7.attn2.to_out.0", "blocks.7.ffn.net.0.proj", "blocks.7.ffn.net.2", "blocks.8.attn1.to_q", "blocks.8.attn1.to_k", "blocks.8.attn1.to_v", "blocks.8.attn1.to_out.0", "blocks.8.attn2.to_q", "blocks.8.attn2.to_k", "blocks.8.attn2.to_v", "blocks.8.attn2.to_out.0", "blocks.8.ffn.net.0.proj", "blocks.8.ffn.net.2", "blocks.9.attn1.to_q", "blocks.9.attn1.to_k", "blocks.9.attn1.to_v", "blocks.9.attn1.to_out.0", "blocks.9.attn2.to_q", "blocks.9.attn2.to_k", "blocks.9.attn2.to_v", "blocks.9.attn2.to_out.0", "blocks.9.ffn.net.0.proj", "blocks.9.ffn.net.2", "blocks.10.attn1.to_q", "blocks.10.attn1.to_k", "blocks.10.attn1.to_v", "blocks.10.attn1.to_out.0", "blocks.10.attn2.to_q", "blocks.10.attn2.to_k", "blocks.10.attn2.to_v", "blocks.10.attn2.to_out.0", "blocks.10.ffn.net.0.proj", "blocks.10.ffn.net.2", "blocks.11.attn1.to_q", "blocks.11.attn1.to_k", "blocks.11.attn1.to_v", "blocks.11.attn1.to_out.0", "blocks.11.attn2.to_q", "blocks.11.attn2.to_k", "blocks.11.attn2.to_v", "blocks.11.attn2.to_out.0", "blocks.11.ffn.net.0.proj", "blocks.11.ffn.net.2", "blocks.12.attn1.to_q", "blocks.12.attn1.to_k", "blocks.12.attn1.to_v", "blocks.12.attn1.to_out.0", "blocks.12.attn2.to_q", "blocks.12.attn2.to_k", "blocks.12.attn2.to_v", "blocks.12.attn2.to_out.0", "blocks.12.ffn.net.0.proj", "blocks.12.ffn.net.2", "blocks.13.attn1.to_q", "blocks.13.attn1.to_k", "blocks.13.attn1.to_v", "blocks.13.attn1.to_out.0", "blocks.13.attn2.to_q", "blocks.13.attn2.to_k", "blocks.13.attn2.to_v", "blocks.13.attn2.to_out.0", "blocks.13.ffn.net.0.proj", "blocks.13.ffn.net.2", "blocks.14.attn1.to_q", "blocks.14.attn1.to_k", "blocks.14.attn1.to_v", "blocks.14.attn1.to_out.0", "blocks.14.attn2.to_q", "blocks.14.attn2.to_k", "blocks.14.attn2.to_v", "blocks.14.attn2.to_out.0", "blocks.14.ffn.net.0.proj", "blocks.14.ffn.net.2", "blocks.15.attn1.to_q", "blocks.15.attn1.to_k", "blocks.15.attn1.to_v", "blocks.15.attn1.to_out.0", "blocks.15.attn2.to_q", "blocks.15.attn2.to_k", "blocks.15.attn2.to_v", "blocks.15.attn2.to_out.0", "blocks.15.ffn.net.0.proj", "blocks.15.ffn.net.2", "blocks.16.attn1.to_q", "blocks.16.attn1.to_k", "blocks.16.attn1.to_v", "blocks.16.attn1.to_out.0", "blocks.16.attn2.to_q", "blocks.16.attn2.to_k", "blocks.16.attn2.to_v", "blocks.16.attn2.to_out.0", "blocks.16.ffn.net.0.proj", "blocks.16.ffn.net.2", "blocks.17.attn1.to_q", "blocks.17.attn1.to_k", "blocks.17.attn1.to_v", "blocks.17.attn1.to_out.0", "blocks.17.attn2.to_q", "blocks.17.attn2.to_k", "blocks.17.attn2.to_v", "blocks.17.attn2.to_out.0", "blocks.17.ffn.net.0.proj", "blocks.17.ffn.net.2", "blocks.18.attn1.to_q", "blocks.18.attn1.to_k", "blocks.18.attn1.to_v", "blocks.18.attn1.to_out.0", "blocks.18.attn2.to_q", "blocks.18.attn2.to_k", "blocks.18.attn2.to_v", "blocks.18.attn2.to_out.0", "blocks.18.ffn.net.0.proj", "blocks.18.ffn.net.2", "blocks.19.attn1.to_q", "blocks.19.attn1.to_k", "blocks.19.attn1.to_v", "blocks.19.attn1.to_out.0", "blocks.19.attn2.to_q", "blocks.19.attn2.to_k", "blocks.19.attn2.to_v", "blocks.19.attn2.to_out.0", "blocks.19.ffn.net.0.proj", "blocks.19.ffn.net.2", "blocks.20.attn1.to_q", "blocks.20.attn1.to_k", "blocks.20.attn1.to_v", "blocks.20.attn1.to_out.0", "blocks.20.attn2.to_q", "blocks.20.attn2.to_k", "blocks.20.attn2.to_v", "blocks.20.attn2.to_out.0", "blocks.20.ffn.net.0.proj", "blocks.20.ffn.net.2", "blocks.21.attn1.to_q", "blocks.21.attn1.to_k", "blocks.21.attn1.to_v", "blocks.21.attn1.to_out.0", "blocks.21.attn2.to_q", "blocks.21.attn2.to_k", "blocks.21.attn2.to_v", "blocks.21.attn2.to_out.0", "blocks.21.ffn.net.0.proj", "blocks.21.ffn.net.2", "blocks.22.attn1.to_q", "blocks.22.attn1.to_k", "blocks.22.attn1.to_v", "blocks.22.attn1.to_out.0", "blocks.22.attn2.to_q", "blocks.22.attn2.to_k", "blocks.22.attn2.to_v", "blocks.22.attn2.to_out.0", "blocks.22.ffn.net.0.proj", "blocks.22.ffn.net.2", "blocks.23.attn1.to_q", "blocks.23.attn1.to_k", "blocks.23.attn1.to_v", "blocks.23.attn1.to_out.0", "blocks.23.attn2.to_q", "blocks.23.attn2.to_k", "blocks.23.attn2.to_v", "blocks.23.attn2.to_out.0", "blocks.23.ffn.net.0.proj", "blocks.23.ffn.net.2", "blocks.24.attn1.to_q", "blocks.24.attn1.to_k", "blocks.24.attn1.to_v", "blocks.24.attn1.to_out.0", "blocks.24.attn2.to_q", "blocks.24.attn2.to_k", "blocks.24.attn2.to_v", "blocks.24.attn2.to_out.0", "blocks.24.ffn.net.0.proj", "blocks.24.ffn.net.2", "blocks.25.attn1.to_q", "blocks.25.attn1.to_k", "blocks.25.attn1.to_v", "blocks.25.attn1.to_out.0", "blocks.25.attn2.to_q", "blocks.25.attn2.to_k", "blocks.25.attn2.to_v", "blocks.25.attn2.to_out.0", "blocks.25.ffn.net.0.proj", "blocks.25.ffn.net.2", "blocks.26.attn1.to_q", "blocks.26.attn1.to_k", "blocks.26.attn1.to_v", "blocks.26.attn1.to_out.0", "blocks.26.attn2.to_q", "blocks.26.attn2.to_k", "blocks.26.attn2.to_v", "blocks.26.attn2.to_out.0", "blocks.26.ffn.net.0.proj", "blocks.26.ffn.net.2", "blocks.27.attn1.to_q", "blocks.27.attn1.to_k", "blocks.27.attn1.to_v", "blocks.27.attn1.to_out.0", "blocks.27.attn2.to_q", "blocks.27.attn2.to_k", "blocks.27.attn2.to_v", "blocks.27.attn2.to_out.0", "blocks.27.ffn.net.0.proj", "blocks.27.ffn.net.2", "blocks.28.attn1.to_q", "blocks.28.attn1.to_k", "blocks.28.attn1.to_v", "blocks.28.attn1.to_out.0", "blocks.28.attn2.to_q", "blocks.28.attn2.to_k", "blocks.28.attn2.to_v", "blocks.28.attn2.to_out.0", "blocks.28.ffn.net.0.proj", "blocks.28.ffn.net.2", "blocks.29.attn1.to_q", "blocks.29.attn1.to_k", "blocks.29.attn1.to_v", "blocks.29.attn1.to_out.0", "blocks.29.attn2.to_q", "blocks.29.attn2.to_k", "blocks.29.attn2.to_v", "blocks.29.attn2.to_out.0", "blocks.29.ffn.net.0.proj", "blocks.29.ffn.net.2" ], "adaln_modules": [], "skipped_modules": [ "condition_embedder.time_embedder.linear_1", "condition_embedder.time_embedder.linear_2", "condition_embedder.time_proj", "condition_embedder.text_embedder.linear_1", "condition_embedder.text_embedder.linear_2", "proj_out" ], "module_shapes": { "scale_shift_table": [ 1, 2, 1536 ], "patch_embedding.weight": [ 1536, 16, 1, 2, 2 ], "patch_embedding.bias": [ 1536 ], "condition_embedder.time_embedder.linear_1.weight": [ 1536, 256 ], "condition_embedder.time_embedder.linear_1.bias": [ 1536 ], "condition_embedder.time_embedder.linear_2.weight": [ 1536, 1536 ], "condition_embedder.time_embedder.linear_2.bias": [ 1536 ], "condition_embedder.time_proj.weight": [ 9216, 1536 ], "condition_embedder.time_proj.bias": [ 9216 ], "condition_embedder.text_embedder.linear_1.weight": [ 1536, 4096 ], "condition_embedder.text_embedder.linear_1.bias": [ 1536 ], "condition_embedder.text_embedder.linear_2.weight": [ 1536, 1536 ], "condition_embedder.text_embedder.linear_2.bias": [ 1536 ], "blocks.0.scale_shift_table": [ 1, 6, 1536 ], "blocks.0.attn1.to_q.bias": [ 1536 ], "blocks.0.attn1.to_q.packed_weight_indices": [ 1179648 ], "blocks.0.attn1.to_q.row_norms": [ 1536 ], "blocks.0.attn1.to_k.bias": [ 1536 ], "blocks.0.attn1.to_k.packed_weight_indices": [ 1179648 ], "blocks.0.attn1.to_k.row_norms": [ 1536 ], "blocks.0.attn1.to_v.bias": [ 1536 ], "blocks.0.attn1.to_v.packed_weight_indices": [ 1179648 ], "blocks.0.attn1.to_v.row_norms": [ 1536 ], "blocks.0.attn1.to_out.0.bias": [ 1536 ], "blocks.0.attn1.to_out.0.packed_weight_indices": [ 1179648 ], "blocks.0.attn1.to_out.0.row_norms": [ 1536 ], "blocks.0.attn1.norm_q.weight": [ 1536 ], "blocks.0.attn1.norm_k.weight": [ 1536 ], "blocks.0.attn2.to_q.bias": [ 1536 ], "blocks.0.attn2.to_q.packed_weight_indices": [ 1179648 ], "blocks.0.attn2.to_q.row_norms": [ 1536 ], "blocks.0.attn2.to_k.bias": [ 1536 ], "blocks.0.attn2.to_k.packed_weight_indices": [ 1179648 ], "blocks.0.attn2.to_k.row_norms": [ 1536 ], "blocks.0.attn2.to_v.bias": [ 1536 ], "blocks.0.attn2.to_v.packed_weight_indices": [ 1179648 ], "blocks.0.attn2.to_v.row_norms": [ 1536 ], "blocks.0.attn2.to_out.0.bias": [ 1536 ], "blocks.0.attn2.to_out.0.packed_weight_indices": [ 1179648 ], "blocks.0.attn2.to_out.0.row_norms": [ 1536 ], "blocks.0.attn2.norm_q.weight": [ 1536 ], "blocks.0.attn2.norm_k.weight": [ 1536 ], "blocks.0.norm2.weight": [ 1536 ], "blocks.0.norm2.bias": [ 1536 ], "blocks.0.ffn.net.0.proj.bias": [ 8960 ], "blocks.0.ffn.net.0.proj.packed_weight_indices": [ 6881280 ], "blocks.0.ffn.net.0.proj.row_norms": [ 8960 ], "blocks.0.ffn.net.2.bias": [ 1536 ], "blocks.0.ffn.net.2.packed_weight_indices": [ 6881280 ], "blocks.0.ffn.net.2.row_norms": [ 1536 ], "blocks.1.scale_shift_table": [ 1, 6, 1536 ], "blocks.1.attn1.to_q.bias": [ 1536 ], "blocks.1.attn1.to_q.packed_weight_indices": [ 1179648 ], "blocks.1.attn1.to_q.row_norms": [ 1536 ], "blocks.1.attn1.to_k.bias": [ 1536 ], "blocks.1.attn1.to_k.packed_weight_indices": [ 1179648 ], "blocks.1.attn1.to_k.row_norms": [ 1536 ], "blocks.1.attn1.to_v.bias": [ 1536 ], "blocks.1.attn1.to_v.packed_weight_indices": [ 1179648 ], "blocks.1.attn1.to_v.row_norms": [ 1536 ], "blocks.1.attn1.to_out.0.bias": [ 1536 ], "blocks.1.attn1.to_out.0.packed_weight_indices": [ 1179648 ], "blocks.1.attn1.to_out.0.row_norms": [ 1536 ], "blocks.1.attn1.norm_q.weight": [ 1536 ], "blocks.1.attn1.norm_k.weight": [ 1536 ], "blocks.1.attn2.to_q.bias": [ 1536 ], "blocks.1.attn2.to_q.packed_weight_indices": [ 1179648 ], "blocks.1.attn2.to_q.row_norms": [ 1536 ], "blocks.1.attn2.to_k.bias": [ 1536 ], "blocks.1.attn2.to_k.packed_weight_indices": [ 1179648 ], "blocks.1.attn2.to_k.row_norms": [ 1536 ], "blocks.1.attn2.to_v.bias": [ 1536 ], "blocks.1.attn2.to_v.packed_weight_indices": [ 1179648 ], "blocks.1.attn2.to_v.row_norms": [ 1536 ], "blocks.1.attn2.to_out.0.bias": [ 1536 ], "blocks.1.attn2.to_out.0.packed_weight_indices": [ 1179648 ], "blocks.1.attn2.to_out.0.row_norms": [ 1536 ], "blocks.1.attn2.norm_q.weight": [ 1536 ], "blocks.1.attn2.norm_k.weight": [ 1536 ], "blocks.1.norm2.weight": [ 1536 ], "blocks.1.norm2.bias": [ 1536 ], "blocks.1.ffn.net.0.proj.bias": [ 8960 ], "blocks.1.ffn.net.0.proj.packed_weight_indices": [ 6881280 ], "blocks.1.ffn.net.0.proj.row_norms": [ 8960 ], "blocks.1.ffn.net.2.bias": [ 1536 ], "blocks.1.ffn.net.2.packed_weight_indices": [ 6881280 ], "blocks.1.ffn.net.2.row_norms": [ 1536 ], "blocks.2.scale_shift_table": [ 1, 6, 1536 ], "blocks.2.attn1.to_q.bias": [ 1536 ], "blocks.2.attn1.to_q.packed_weight_indices": [ 1179648 ], "blocks.2.attn1.to_q.row_norms": [ 1536 ], "blocks.2.attn1.to_k.bias": [ 1536 ], "blocks.2.attn1.to_k.packed_weight_indices": [ 1179648 ], "blocks.2.attn1.to_k.row_norms": [ 1536 ], "blocks.2.attn1.to_v.bias": [ 1536 ], "blocks.2.attn1.to_v.packed_weight_indices": [ 1179648 ], "blocks.2.attn1.to_v.row_norms": [ 1536 ], "blocks.2.attn1.to_out.0.bias": [ 1536 ], "blocks.2.attn1.to_out.0.packed_weight_indices": [ 1179648 ], "blocks.2.attn1.to_out.0.row_norms": [ 1536 ], "blocks.2.attn1.norm_q.weight": [ 1536 ], "blocks.2.attn1.norm_k.weight": [ 1536 ], "blocks.2.attn2.to_q.bias": [ 1536 ], "blocks.2.attn2.to_q.packed_weight_indices": [ 1179648 ], "blocks.2.attn2.to_q.row_norms": [ 1536 ], "blocks.2.attn2.to_k.bias": [ 1536 ], "blocks.2.attn2.to_k.packed_weight_indices": [ 1179648 ], "blocks.2.attn2.to_k.row_norms": [ 1536 ], "blocks.2.attn2.to_v.bias": [ 1536 ], "blocks.2.attn2.to_v.packed_weight_indices": [ 1179648 ], "blocks.2.attn2.to_v.row_norms": [ 1536 ], "blocks.2.attn2.to_out.0.bias": [ 1536 ], "blocks.2.attn2.to_out.0.packed_weight_indices": [ 1179648 ], "blocks.2.attn2.to_out.0.row_norms": [ 1536 ], "blocks.2.attn2.norm_q.weight": [ 1536 ], "blocks.2.attn2.norm_k.weight": [ 1536 ], "blocks.2.norm2.weight": [ 1536 ], "blocks.2.norm2.bias": [ 1536 ], "blocks.2.ffn.net.0.proj.bias": [ 8960 ], "blocks.2.ffn.net.0.proj.packed_weight_indices": [ 6881280 ], "blocks.2.ffn.net.0.proj.row_norms": [ 8960 ], "blocks.2.ffn.net.2.bias": [ 1536 ], "blocks.2.ffn.net.2.packed_weight_indices": [ 6881280 ], "blocks.2.ffn.net.2.row_norms": [ 1536 ], "blocks.3.scale_shift_table": [ 1, 6, 1536 ], "blocks.3.attn1.to_q.bias": [ 1536 ], "blocks.3.attn1.to_q.packed_weight_indices": [ 1179648 ], "blocks.3.attn1.to_q.row_norms": [ 1536 ], "blocks.3.attn1.to_k.bias": [ 1536 ], "blocks.3.attn1.to_k.packed_weight_indices": [ 1179648 ], "blocks.3.attn1.to_k.row_norms": [ 1536 ], "blocks.3.attn1.to_v.bias": [ 1536 ], "blocks.3.attn1.to_v.packed_weight_indices": [ 1179648 ], "blocks.3.attn1.to_v.row_norms": [ 1536 ], "blocks.3.attn1.to_out.0.bias": [ 1536 ], "blocks.3.attn1.to_out.0.packed_weight_indices": [ 1179648 ], "blocks.3.attn1.to_out.0.row_norms": [ 1536 ], "blocks.3.attn1.norm_q.weight": [ 1536 ], "blocks.3.attn1.norm_k.weight": [ 1536 ], "blocks.3.attn2.to_q.bias": [ 1536 ], "blocks.3.attn2.to_q.packed_weight_indices": [ 1179648 ], "blocks.3.attn2.to_q.row_norms": [ 1536 ], "blocks.3.attn2.to_k.bias": [ 1536 ], "blocks.3.attn2.to_k.packed_weight_indices": [ 1179648 ], "blocks.3.attn2.to_k.row_norms": [ 1536 ], "blocks.3.attn2.to_v.bias": [ 1536 ], "blocks.3.attn2.to_v.packed_weight_indices": [ 1179648 ], "blocks.3.attn2.to_v.row_norms": [ 1536 ], "blocks.3.attn2.to_out.0.bias": [ 1536 ], "blocks.3.attn2.to_out.0.packed_weight_indices": [ 1179648 ], "blocks.3.attn2.to_out.0.row_norms": [ 1536 ], "blocks.3.attn2.norm_q.weight": [ 1536 ], "blocks.3.attn2.norm_k.weight": [ 1536 ], "blocks.3.norm2.weight": [ 1536 ], "blocks.3.norm2.bias": [ 1536 ], "blocks.3.ffn.net.0.proj.bias": [ 8960 ], "blocks.3.ffn.net.0.proj.packed_weight_indices": [ 6881280 ], "blocks.3.ffn.net.0.proj.row_norms": [ 8960 ], "blocks.3.ffn.net.2.bias": [ 1536 ], "blocks.3.ffn.net.2.packed_weight_indices": [ 6881280 ], "blocks.3.ffn.net.2.row_norms": [ 1536 ], "blocks.4.scale_shift_table": [ 1, 6, 1536 ], "blocks.4.attn1.to_q.bias": [ 1536 ], "blocks.4.attn1.to_q.packed_weight_indices": [ 1179648 ], "blocks.4.attn1.to_q.row_norms": [ 1536 ], "blocks.4.attn1.to_k.bias": [ 1536 ], "blocks.4.attn1.to_k.packed_weight_indices": [ 1179648 ], "blocks.4.attn1.to_k.row_norms": [ 1536 ], "blocks.4.attn1.to_v.bias": [ 1536 ], "blocks.4.attn1.to_v.packed_weight_indices": [ 1179648 ], "blocks.4.attn1.to_v.row_norms": [ 1536 ], "blocks.4.attn1.to_out.0.bias": [ 1536 ], "blocks.4.attn1.to_out.0.packed_weight_indices": [ 1179648 ], "blocks.4.attn1.to_out.0.row_norms": [ 1536 ], "blocks.4.attn1.norm_q.weight": [ 1536 ], "blocks.4.attn1.norm_k.weight": [ 1536 ], "blocks.4.attn2.to_q.bias": [ 1536 ], "blocks.4.attn2.to_q.packed_weight_indices": [ 1179648 ], "blocks.4.attn2.to_q.row_norms": [ 1536 ], "blocks.4.attn2.to_k.bias": [ 1536 ], "blocks.4.attn2.to_k.packed_weight_indices": [ 1179648 ], "blocks.4.attn2.to_k.row_norms": [ 1536 ], "blocks.4.attn2.to_v.bias": [ 1536 ], "blocks.4.attn2.to_v.packed_weight_indices": [ 1179648 ], "blocks.4.attn2.to_v.row_norms": [ 1536 ], "blocks.4.attn2.to_out.0.bias": [ 1536 ], "blocks.4.attn2.to_out.0.packed_weight_indices": [ 1179648 ], "blocks.4.attn2.to_out.0.row_norms": [ 1536 ], "blocks.4.attn2.norm_q.weight": [ 1536 ], "blocks.4.attn2.norm_k.weight": [ 1536 ], "blocks.4.norm2.weight": [ 1536 ], "blocks.4.norm2.bias": [ 1536 ], "blocks.4.ffn.net.0.proj.bias": [ 8960 ], "blocks.4.ffn.net.0.proj.packed_weight_indices": [ 6881280 ], "blocks.4.ffn.net.0.proj.row_norms": [ 8960 ], "blocks.4.ffn.net.2.bias": [ 1536 ], "blocks.4.ffn.net.2.packed_weight_indices": [ 6881280 ], "blocks.4.ffn.net.2.row_norms": [ 1536 ], "blocks.5.scale_shift_table": [ 1, 6, 1536 ], "blocks.5.attn1.to_q.bias": [ 1536 ], "blocks.5.attn1.to_q.packed_weight_indices": [ 1179648 ], "blocks.5.attn1.to_q.row_norms": [ 1536 ], "blocks.5.attn1.to_k.bias": [ 1536 ], "blocks.5.attn1.to_k.packed_weight_indices": [ 1179648 ], "blocks.5.attn1.to_k.row_norms": [ 1536 ], "blocks.5.attn1.to_v.bias": [ 1536 ], "blocks.5.attn1.to_v.packed_weight_indices": [ 1179648 ], "blocks.5.attn1.to_v.row_norms": [ 1536 ], "blocks.5.attn1.to_out.0.bias": [ 1536 ], "blocks.5.attn1.to_out.0.packed_weight_indices": [ 1179648 ], "blocks.5.attn1.to_out.0.row_norms": [ 1536 ], "blocks.5.attn1.norm_q.weight": [ 1536 ], "blocks.5.attn1.norm_k.weight": [ 1536 ], "blocks.5.attn2.to_q.bias": [ 1536 ], "blocks.5.attn2.to_q.packed_weight_indices": [ 1179648 ], "blocks.5.attn2.to_q.row_norms": [ 1536 ], "blocks.5.attn2.to_k.bias": [ 1536 ], "blocks.5.attn2.to_k.packed_weight_indices": [ 1179648 ], "blocks.5.attn2.to_k.row_norms": [ 1536 ], "blocks.5.attn2.to_v.bias": [ 1536 ], "blocks.5.attn2.to_v.packed_weight_indices": [ 1179648 ], "blocks.5.attn2.to_v.row_norms": [ 1536 ], "blocks.5.attn2.to_out.0.bias": [ 1536 ], "blocks.5.attn2.to_out.0.packed_weight_indices": [ 1179648 ], "blocks.5.attn2.to_out.0.row_norms": [ 1536 ], "blocks.5.attn2.norm_q.weight": [ 1536 ], "blocks.5.attn2.norm_k.weight": [ 1536 ], "blocks.5.norm2.weight": [ 1536 ], "blocks.5.norm2.bias": [ 1536 ], "blocks.5.ffn.net.0.proj.bias": [ 8960 ], "blocks.5.ffn.net.0.proj.packed_weight_indices": [ 6881280 ], "blocks.5.ffn.net.0.proj.row_norms": [ 8960 ], "blocks.5.ffn.net.2.bias": [ 1536 ], "blocks.5.ffn.net.2.packed_weight_indices": [ 6881280 ], "blocks.5.ffn.net.2.row_norms": [ 1536 ], "blocks.6.scale_shift_table": [ 1, 6, 1536 ], "blocks.6.attn1.to_q.bias": [ 1536 ], "blocks.6.attn1.to_q.packed_weight_indices": [ 1179648 ], "blocks.6.attn1.to_q.row_norms": [ 1536 ], "blocks.6.attn1.to_k.bias": [ 1536 ], "blocks.6.attn1.to_k.packed_weight_indices": [ 1179648 ], "blocks.6.attn1.to_k.row_norms": [ 1536 ], "blocks.6.attn1.to_v.bias": [ 1536 ], "blocks.6.attn1.to_v.packed_weight_indices": [ 1179648 ], "blocks.6.attn1.to_v.row_norms": [ 1536 ], "blocks.6.attn1.to_out.0.bias": [ 1536 ], "blocks.6.attn1.to_out.0.packed_weight_indices": [ 1179648 ], "blocks.6.attn1.to_out.0.row_norms": [ 1536 ], "blocks.6.attn1.norm_q.weight": [ 1536 ], "blocks.6.attn1.norm_k.weight": [ 1536 ], "blocks.6.attn2.to_q.bias": [ 1536 ], "blocks.6.attn2.to_q.packed_weight_indices": [ 1179648 ], "blocks.6.attn2.to_q.row_norms": [ 1536 ], "blocks.6.attn2.to_k.bias": [ 1536 ], "blocks.6.attn2.to_k.packed_weight_indices": [ 1179648 ], "blocks.6.attn2.to_k.row_norms": [ 1536 ], "blocks.6.attn2.to_v.bias": [ 1536 ], "blocks.6.attn2.to_v.packed_weight_indices": [ 1179648 ], "blocks.6.attn2.to_v.row_norms": [ 1536 ], "blocks.6.attn2.to_out.0.bias": [ 1536 ], "blocks.6.attn2.to_out.0.packed_weight_indices": [ 1179648 ], "blocks.6.attn2.to_out.0.row_norms": [ 1536 ], "blocks.6.attn2.norm_q.weight": [ 1536 ], "blocks.6.attn2.norm_k.weight": [ 1536 ], "blocks.6.norm2.weight": [ 1536 ], "blocks.6.norm2.bias": [ 1536 ], "blocks.6.ffn.net.0.proj.bias": [ 8960 ], "blocks.6.ffn.net.0.proj.packed_weight_indices": [ 6881280 ], "blocks.6.ffn.net.0.proj.row_norms": [ 8960 ], "blocks.6.ffn.net.2.bias": [ 1536 ], "blocks.6.ffn.net.2.packed_weight_indices": [ 6881280 ], "blocks.6.ffn.net.2.row_norms": [ 1536 ], "blocks.7.scale_shift_table": [ 1, 6, 1536 ], "blocks.7.attn1.to_q.bias": [ 1536 ], "blocks.7.attn1.to_q.packed_weight_indices": [ 1179648 ], "blocks.7.attn1.to_q.row_norms": [ 1536 ], "blocks.7.attn1.to_k.bias": [ 1536 ], "blocks.7.attn1.to_k.packed_weight_indices": [ 1179648 ], "blocks.7.attn1.to_k.row_norms": [ 1536 ], "blocks.7.attn1.to_v.bias": [ 1536 ], "blocks.7.attn1.to_v.packed_weight_indices": [ 1179648 ], "blocks.7.attn1.to_v.row_norms": [ 1536 ], "blocks.7.attn1.to_out.0.bias": [ 1536 ], "blocks.7.attn1.to_out.0.packed_weight_indices": [ 1179648 ], "blocks.7.attn1.to_out.0.row_norms": [ 1536 ], "blocks.7.attn1.norm_q.weight": [ 1536 ], "blocks.7.attn1.norm_k.weight": [ 1536 ], "blocks.7.attn2.to_q.bias": [ 1536 ], "blocks.7.attn2.to_q.packed_weight_indices": [ 1179648 ], "blocks.7.attn2.to_q.row_norms": [ 1536 ], "blocks.7.attn2.to_k.bias": [ 1536 ], "blocks.7.attn2.to_k.packed_weight_indices": [ 1179648 ], "blocks.7.attn2.to_k.row_norms": [ 1536 ], "blocks.7.attn2.to_v.bias": [ 1536 ], "blocks.7.attn2.to_v.packed_weight_indices": [ 1179648 ], "blocks.7.attn2.to_v.row_norms": [ 1536 ], "blocks.7.attn2.to_out.0.bias": [ 1536 ], "blocks.7.attn2.to_out.0.packed_weight_indices": [ 1179648 ], "blocks.7.attn2.to_out.0.row_norms": [ 1536 ], "blocks.7.attn2.norm_q.weight": [ 1536 ], "blocks.7.attn2.norm_k.weight": [ 1536 ], "blocks.7.norm2.weight": [ 1536 ], "blocks.7.norm2.bias": [ 1536 ], "blocks.7.ffn.net.0.proj.bias": [ 8960 ], "blocks.7.ffn.net.0.proj.packed_weight_indices": [ 6881280 ], "blocks.7.ffn.net.0.proj.row_norms": [ 8960 ], "blocks.7.ffn.net.2.bias": [ 1536 ], "blocks.7.ffn.net.2.packed_weight_indices": [ 6881280 ], "blocks.7.ffn.net.2.row_norms": [ 1536 ], "blocks.8.scale_shift_table": [ 1, 6, 1536 ], "blocks.8.attn1.to_q.bias": [ 1536 ], "blocks.8.attn1.to_q.packed_weight_indices": [ 1179648 ], "blocks.8.attn1.to_q.row_norms": [ 1536 ], "blocks.8.attn1.to_k.bias": [ 1536 ], "blocks.8.attn1.to_k.packed_weight_indices": [ 1179648 ], "blocks.8.attn1.to_k.row_norms": [ 1536 ], "blocks.8.attn1.to_v.bias": [ 1536 ], "blocks.8.attn1.to_v.packed_weight_indices": [ 1179648 ], "blocks.8.attn1.to_v.row_norms": [ 1536 ], "blocks.8.attn1.to_out.0.bias": [ 1536 ], "blocks.8.attn1.to_out.0.packed_weight_indices": [ 1179648 ], "blocks.8.attn1.to_out.0.row_norms": [ 1536 ], "blocks.8.attn1.norm_q.weight": [ 1536 ], "blocks.8.attn1.norm_k.weight": [ 1536 ], "blocks.8.attn2.to_q.bias": [ 1536 ], "blocks.8.attn2.to_q.packed_weight_indices": [ 1179648 ], "blocks.8.attn2.to_q.row_norms": [ 1536 ], "blocks.8.attn2.to_k.bias": [ 1536 ], "blocks.8.attn2.to_k.packed_weight_indices": [ 1179648 ], "blocks.8.attn2.to_k.row_norms": [ 1536 ], "blocks.8.attn2.to_v.bias": [ 1536 ], "blocks.8.attn2.to_v.packed_weight_indices": [ 1179648 ], "blocks.8.attn2.to_v.row_norms": [ 1536 ], "blocks.8.attn2.to_out.0.bias": [ 1536 ], "blocks.8.attn2.to_out.0.packed_weight_indices": [ 1179648 ], "blocks.8.attn2.to_out.0.row_norms": [ 1536 ], "blocks.8.attn2.norm_q.weight": [ 1536 ], "blocks.8.attn2.norm_k.weight": [ 1536 ], "blocks.8.norm2.weight": [ 1536 ], "blocks.8.norm2.bias": [ 1536 ], "blocks.8.ffn.net.0.proj.bias": [ 8960 ], "blocks.8.ffn.net.0.proj.packed_weight_indices": [ 6881280 ], "blocks.8.ffn.net.0.proj.row_norms": [ 8960 ], "blocks.8.ffn.net.2.bias": [ 1536 ], "blocks.8.ffn.net.2.packed_weight_indices": [ 6881280 ], "blocks.8.ffn.net.2.row_norms": [ 1536 ], "blocks.9.scale_shift_table": [ 1, 6, 1536 ], "blocks.9.attn1.to_q.bias": [ 1536 ], "blocks.9.attn1.to_q.packed_weight_indices": [ 1179648 ], "blocks.9.attn1.to_q.row_norms": [ 1536 ], "blocks.9.attn1.to_k.bias": [ 1536 ], "blocks.9.attn1.to_k.packed_weight_indices": [ 1179648 ], "blocks.9.attn1.to_k.row_norms": [ 1536 ], "blocks.9.attn1.to_v.bias": [ 1536 ], "blocks.9.attn1.to_v.packed_weight_indices": [ 1179648 ], "blocks.9.attn1.to_v.row_norms": [ 1536 ], "blocks.9.attn1.to_out.0.bias": [ 1536 ], "blocks.9.attn1.to_out.0.packed_weight_indices": [ 1179648 ], "blocks.9.attn1.to_out.0.row_norms": [ 1536 ], "blocks.9.attn1.norm_q.weight": [ 1536 ], "blocks.9.attn1.norm_k.weight": [ 1536 ], "blocks.9.attn2.to_q.bias": [ 1536 ], "blocks.9.attn2.to_q.packed_weight_indices": [ 1179648 ], "blocks.9.attn2.to_q.row_norms": [ 1536 ], "blocks.9.attn2.to_k.bias": [ 1536 ], "blocks.9.attn2.to_k.packed_weight_indices": [ 1179648 ], "blocks.9.attn2.to_k.row_norms": [ 1536 ], "blocks.9.attn2.to_v.bias": [ 1536 ], "blocks.9.attn2.to_v.packed_weight_indices": [ 1179648 ], "blocks.9.attn2.to_v.row_norms": [ 1536 ], "blocks.9.attn2.to_out.0.bias": [ 1536 ], "blocks.9.attn2.to_out.0.packed_weight_indices": [ 1179648 ], "blocks.9.attn2.to_out.0.row_norms": [ 1536 ], "blocks.9.attn2.norm_q.weight": [ 1536 ], "blocks.9.attn2.norm_k.weight": [ 1536 ], "blocks.9.norm2.weight": [ 1536 ], "blocks.9.norm2.bias": [ 1536 ], "blocks.9.ffn.net.0.proj.bias": [ 8960 ], "blocks.9.ffn.net.0.proj.packed_weight_indices": [ 6881280 ], "blocks.9.ffn.net.0.proj.row_norms": [ 8960 ], "blocks.9.ffn.net.2.bias": [ 1536 ], "blocks.9.ffn.net.2.packed_weight_indices": [ 6881280 ], "blocks.9.ffn.net.2.row_norms": [ 1536 ], "blocks.10.scale_shift_table": [ 1, 6, 1536 ], "blocks.10.attn1.to_q.bias": [ 1536 ], "blocks.10.attn1.to_q.packed_weight_indices": [ 1179648 ], "blocks.10.attn1.to_q.row_norms": [ 1536 ], "blocks.10.attn1.to_k.bias": [ 1536 ], "blocks.10.attn1.to_k.packed_weight_indices": [ 1179648 ], "blocks.10.attn1.to_k.row_norms": [ 1536 ], "blocks.10.attn1.to_v.bias": [ 1536 ], "blocks.10.attn1.to_v.packed_weight_indices": [ 1179648 ], "blocks.10.attn1.to_v.row_norms": [ 1536 ], "blocks.10.attn1.to_out.0.bias": [ 1536 ], "blocks.10.attn1.to_out.0.packed_weight_indices": [ 1179648 ], "blocks.10.attn1.to_out.0.row_norms": [ 1536 ], "blocks.10.attn1.norm_q.weight": [ 1536 ], "blocks.10.attn1.norm_k.weight": [ 1536 ], "blocks.10.attn2.to_q.bias": [ 1536 ], "blocks.10.attn2.to_q.packed_weight_indices": [ 1179648 ], "blocks.10.attn2.to_q.row_norms": [ 1536 ], "blocks.10.attn2.to_k.bias": [ 1536 ], "blocks.10.attn2.to_k.packed_weight_indices": [ 1179648 ], "blocks.10.attn2.to_k.row_norms": [ 1536 ], "blocks.10.attn2.to_v.bias": [ 1536 ], "blocks.10.attn2.to_v.packed_weight_indices": [ 1179648 ], "blocks.10.attn2.to_v.row_norms": [ 1536 ], "blocks.10.attn2.to_out.0.bias": [ 1536 ], "blocks.10.attn2.to_out.0.packed_weight_indices": [ 1179648 ], "blocks.10.attn2.to_out.0.row_norms": [ 1536 ], "blocks.10.attn2.norm_q.weight": [ 1536 ], "blocks.10.attn2.norm_k.weight": [ 1536 ], "blocks.10.norm2.weight": [ 1536 ], "blocks.10.norm2.bias": [ 1536 ], "blocks.10.ffn.net.0.proj.bias": [ 8960 ], "blocks.10.ffn.net.0.proj.packed_weight_indices": [ 6881280 ], "blocks.10.ffn.net.0.proj.row_norms": [ 8960 ], "blocks.10.ffn.net.2.bias": [ 1536 ], "blocks.10.ffn.net.2.packed_weight_indices": [ 6881280 ], "blocks.10.ffn.net.2.row_norms": [ 1536 ], "blocks.11.scale_shift_table": [ 1, 6, 1536 ], "blocks.11.attn1.to_q.bias": [ 1536 ], "blocks.11.attn1.to_q.packed_weight_indices": [ 1179648 ], "blocks.11.attn1.to_q.row_norms": [ 1536 ], "blocks.11.attn1.to_k.bias": [ 1536 ], "blocks.11.attn1.to_k.packed_weight_indices": [ 1179648 ], "blocks.11.attn1.to_k.row_norms": [ 1536 ], "blocks.11.attn1.to_v.bias": [ 1536 ], "blocks.11.attn1.to_v.packed_weight_indices": [ 1179648 ], "blocks.11.attn1.to_v.row_norms": [ 1536 ], "blocks.11.attn1.to_out.0.bias": [ 1536 ], "blocks.11.attn1.to_out.0.packed_weight_indices": [ 1179648 ], "blocks.11.attn1.to_out.0.row_norms": [ 1536 ], "blocks.11.attn1.norm_q.weight": [ 1536 ], "blocks.11.attn1.norm_k.weight": [ 1536 ], "blocks.11.attn2.to_q.bias": [ 1536 ], "blocks.11.attn2.to_q.packed_weight_indices": [ 1179648 ], "blocks.11.attn2.to_q.row_norms": [ 1536 ], "blocks.11.attn2.to_k.bias": [ 1536 ], "blocks.11.attn2.to_k.packed_weight_indices": [ 1179648 ], "blocks.11.attn2.to_k.row_norms": [ 1536 ], "blocks.11.attn2.to_v.bias": [ 1536 ], "blocks.11.attn2.to_v.packed_weight_indices": [ 1179648 ], "blocks.11.attn2.to_v.row_norms": [ 1536 ], "blocks.11.attn2.to_out.0.bias": [ 1536 ], "blocks.11.attn2.to_out.0.packed_weight_indices": [ 1179648 ], "blocks.11.attn2.to_out.0.row_norms": [ 1536 ], "blocks.11.attn2.norm_q.weight": [ 1536 ], "blocks.11.attn2.norm_k.weight": [ 1536 ], "blocks.11.norm2.weight": [ 1536 ], "blocks.11.norm2.bias": [ 1536 ], "blocks.11.ffn.net.0.proj.bias": [ 8960 ], "blocks.11.ffn.net.0.proj.packed_weight_indices": [ 6881280 ], "blocks.11.ffn.net.0.proj.row_norms": [ 8960 ], "blocks.11.ffn.net.2.bias": [ 1536 ], "blocks.11.ffn.net.2.packed_weight_indices": [ 6881280 ], "blocks.11.ffn.net.2.row_norms": [ 1536 ], "blocks.12.scale_shift_table": [ 1, 6, 1536 ], "blocks.12.attn1.to_q.bias": [ 1536 ], "blocks.12.attn1.to_q.packed_weight_indices": [ 1179648 ], "blocks.12.attn1.to_q.row_norms": [ 1536 ], "blocks.12.attn1.to_k.bias": [ 1536 ], "blocks.12.attn1.to_k.packed_weight_indices": [ 1179648 ], "blocks.12.attn1.to_k.row_norms": [ 1536 ], "blocks.12.attn1.to_v.bias": [ 1536 ], "blocks.12.attn1.to_v.packed_weight_indices": [ 1179648 ], "blocks.12.attn1.to_v.row_norms": [ 1536 ], "blocks.12.attn1.to_out.0.bias": [ 1536 ], "blocks.12.attn1.to_out.0.packed_weight_indices": [ 1179648 ], "blocks.12.attn1.to_out.0.row_norms": [ 1536 ], "blocks.12.attn1.norm_q.weight": [ 1536 ], "blocks.12.attn1.norm_k.weight": [ 1536 ], "blocks.12.attn2.to_q.bias": [ 1536 ], "blocks.12.attn2.to_q.packed_weight_indices": [ 1179648 ], "blocks.12.attn2.to_q.row_norms": [ 1536 ], "blocks.12.attn2.to_k.bias": [ 1536 ], "blocks.12.attn2.to_k.packed_weight_indices": [ 1179648 ], "blocks.12.attn2.to_k.row_norms": [ 1536 ], "blocks.12.attn2.to_v.bias": [ 1536 ], "blocks.12.attn2.to_v.packed_weight_indices": [ 1179648 ], "blocks.12.attn2.to_v.row_norms": [ 1536 ], "blocks.12.attn2.to_out.0.bias": [ 1536 ], "blocks.12.attn2.to_out.0.packed_weight_indices": [ 1179648 ], "blocks.12.attn2.to_out.0.row_norms": [ 1536 ], "blocks.12.attn2.norm_q.weight": [ 1536 ], "blocks.12.attn2.norm_k.weight": [ 1536 ], "blocks.12.norm2.weight": [ 1536 ], "blocks.12.norm2.bias": [ 1536 ], "blocks.12.ffn.net.0.proj.bias": [ 8960 ], "blocks.12.ffn.net.0.proj.packed_weight_indices": [ 6881280 ], "blocks.12.ffn.net.0.proj.row_norms": [ 8960 ], "blocks.12.ffn.net.2.bias": [ 1536 ], "blocks.12.ffn.net.2.packed_weight_indices": [ 6881280 ], "blocks.12.ffn.net.2.row_norms": [ 1536 ], "blocks.13.scale_shift_table": [ 1, 6, 1536 ], "blocks.13.attn1.to_q.bias": [ 1536 ], "blocks.13.attn1.to_q.packed_weight_indices": [ 1179648 ], "blocks.13.attn1.to_q.row_norms": [ 1536 ], "blocks.13.attn1.to_k.bias": [ 1536 ], "blocks.13.attn1.to_k.packed_weight_indices": [ 1179648 ], "blocks.13.attn1.to_k.row_norms": [ 1536 ], "blocks.13.attn1.to_v.bias": [ 1536 ], "blocks.13.attn1.to_v.packed_weight_indices": [ 1179648 ], "blocks.13.attn1.to_v.row_norms": [ 1536 ], "blocks.13.attn1.to_out.0.bias": [ 1536 ], "blocks.13.attn1.to_out.0.packed_weight_indices": [ 1179648 ], "blocks.13.attn1.to_out.0.row_norms": [ 1536 ], "blocks.13.attn1.norm_q.weight": [ 1536 ], "blocks.13.attn1.norm_k.weight": [ 1536 ], "blocks.13.attn2.to_q.bias": [ 1536 ], "blocks.13.attn2.to_q.packed_weight_indices": [ 1179648 ], "blocks.13.attn2.to_q.row_norms": [ 1536 ], "blocks.13.attn2.to_k.bias": [ 1536 ], "blocks.13.attn2.to_k.packed_weight_indices": [ 1179648 ], "blocks.13.attn2.to_k.row_norms": [ 1536 ], "blocks.13.attn2.to_v.bias": [ 1536 ], "blocks.13.attn2.to_v.packed_weight_indices": [ 1179648 ], "blocks.13.attn2.to_v.row_norms": [ 1536 ], "blocks.13.attn2.to_out.0.bias": [ 1536 ], "blocks.13.attn2.to_out.0.packed_weight_indices": [ 1179648 ], "blocks.13.attn2.to_out.0.row_norms": [ 1536 ], "blocks.13.attn2.norm_q.weight": [ 1536 ], "blocks.13.attn2.norm_k.weight": [ 1536 ], "blocks.13.norm2.weight": [ 1536 ], "blocks.13.norm2.bias": [ 1536 ], "blocks.13.ffn.net.0.proj.bias": [ 8960 ], "blocks.13.ffn.net.0.proj.packed_weight_indices": [ 6881280 ], "blocks.13.ffn.net.0.proj.row_norms": [ 8960 ], "blocks.13.ffn.net.2.bias": [ 1536 ], "blocks.13.ffn.net.2.packed_weight_indices": [ 6881280 ], "blocks.13.ffn.net.2.row_norms": [ 1536 ], "blocks.14.scale_shift_table": [ 1, 6, 1536 ], "blocks.14.attn1.to_q.bias": [ 1536 ], "blocks.14.attn1.to_q.packed_weight_indices": [ 1179648 ], "blocks.14.attn1.to_q.row_norms": [ 1536 ], "blocks.14.attn1.to_k.bias": [ 1536 ], "blocks.14.attn1.to_k.packed_weight_indices": [ 1179648 ], "blocks.14.attn1.to_k.row_norms": [ 1536 ], "blocks.14.attn1.to_v.bias": [ 1536 ], "blocks.14.attn1.to_v.packed_weight_indices": [ 1179648 ], "blocks.14.attn1.to_v.row_norms": [ 1536 ], "blocks.14.attn1.to_out.0.bias": [ 1536 ], "blocks.14.attn1.to_out.0.packed_weight_indices": [ 1179648 ], "blocks.14.attn1.to_out.0.row_norms": [ 1536 ], "blocks.14.attn1.norm_q.weight": [ 1536 ], "blocks.14.attn1.norm_k.weight": [ 1536 ], "blocks.14.attn2.to_q.bias": [ 1536 ], "blocks.14.attn2.to_q.packed_weight_indices": [ 1179648 ], "blocks.14.attn2.to_q.row_norms": [ 1536 ], "blocks.14.attn2.to_k.bias": [ 1536 ], "blocks.14.attn2.to_k.packed_weight_indices": [ 1179648 ], "blocks.14.attn2.to_k.row_norms": [ 1536 ], "blocks.14.attn2.to_v.bias": [ 1536 ], "blocks.14.attn2.to_v.packed_weight_indices": [ 1179648 ], "blocks.14.attn2.to_v.row_norms": [ 1536 ], "blocks.14.attn2.to_out.0.bias": [ 1536 ], "blocks.14.attn2.to_out.0.packed_weight_indices": [ 1179648 ], "blocks.14.attn2.to_out.0.row_norms": [ 1536 ], "blocks.14.attn2.norm_q.weight": [ 1536 ], "blocks.14.attn2.norm_k.weight": [ 1536 ], "blocks.14.norm2.weight": [ 1536 ], "blocks.14.norm2.bias": [ 1536 ], "blocks.14.ffn.net.0.proj.bias": [ 8960 ], "blocks.14.ffn.net.0.proj.packed_weight_indices": [ 6881280 ], "blocks.14.ffn.net.0.proj.row_norms": [ 8960 ], "blocks.14.ffn.net.2.bias": [ 1536 ], "blocks.14.ffn.net.2.packed_weight_indices": [ 6881280 ], "blocks.14.ffn.net.2.row_norms": [ 1536 ], "blocks.15.scale_shift_table": [ 1, 6, 1536 ], "blocks.15.attn1.to_q.bias": [ 1536 ], "blocks.15.attn1.to_q.packed_weight_indices": [ 1179648 ], "blocks.15.attn1.to_q.row_norms": [ 1536 ], "blocks.15.attn1.to_k.bias": [ 1536 ], "blocks.15.attn1.to_k.packed_weight_indices": [ 1179648 ], "blocks.15.attn1.to_k.row_norms": [ 1536 ], "blocks.15.attn1.to_v.bias": [ 1536 ], "blocks.15.attn1.to_v.packed_weight_indices": [ 1179648 ], "blocks.15.attn1.to_v.row_norms": [ 1536 ], "blocks.15.attn1.to_out.0.bias": [ 1536 ], "blocks.15.attn1.to_out.0.packed_weight_indices": [ 1179648 ], "blocks.15.attn1.to_out.0.row_norms": [ 1536 ], "blocks.15.attn1.norm_q.weight": [ 1536 ], "blocks.15.attn1.norm_k.weight": [ 1536 ], "blocks.15.attn2.to_q.bias": [ 1536 ], "blocks.15.attn2.to_q.packed_weight_indices": [ 1179648 ], "blocks.15.attn2.to_q.row_norms": [ 1536 ], "blocks.15.attn2.to_k.bias": [ 1536 ], "blocks.15.attn2.to_k.packed_weight_indices": [ 1179648 ], "blocks.15.attn2.to_k.row_norms": [ 1536 ], "blocks.15.attn2.to_v.bias": [ 1536 ], "blocks.15.attn2.to_v.packed_weight_indices": [ 1179648 ], "blocks.15.attn2.to_v.row_norms": [ 1536 ], "blocks.15.attn2.to_out.0.bias": [ 1536 ], "blocks.15.attn2.to_out.0.packed_weight_indices": [ 1179648 ], "blocks.15.attn2.to_out.0.row_norms": [ 1536 ], "blocks.15.attn2.norm_q.weight": [ 1536 ], "blocks.15.attn2.norm_k.weight": [ 1536 ], "blocks.15.norm2.weight": [ 1536 ], "blocks.15.norm2.bias": [ 1536 ], "blocks.15.ffn.net.0.proj.bias": [ 8960 ], "blocks.15.ffn.net.0.proj.packed_weight_indices": [ 6881280 ], "blocks.15.ffn.net.0.proj.row_norms": [ 8960 ], "blocks.15.ffn.net.2.bias": [ 1536 ], "blocks.15.ffn.net.2.packed_weight_indices": [ 6881280 ], "blocks.15.ffn.net.2.row_norms": [ 1536 ], "blocks.16.scale_shift_table": [ 1, 6, 1536 ], "blocks.16.attn1.to_q.bias": [ 1536 ], "blocks.16.attn1.to_q.packed_weight_indices": [ 1179648 ], "blocks.16.attn1.to_q.row_norms": [ 1536 ], "blocks.16.attn1.to_k.bias": [ 1536 ], "blocks.16.attn1.to_k.packed_weight_indices": [ 1179648 ], "blocks.16.attn1.to_k.row_norms": [ 1536 ], "blocks.16.attn1.to_v.bias": [ 1536 ], "blocks.16.attn1.to_v.packed_weight_indices": [ 1179648 ], "blocks.16.attn1.to_v.row_norms": [ 1536 ], "blocks.16.attn1.to_out.0.bias": [ 1536 ], "blocks.16.attn1.to_out.0.packed_weight_indices": [ 1179648 ], "blocks.16.attn1.to_out.0.row_norms": [ 1536 ], "blocks.16.attn1.norm_q.weight": [ 1536 ], "blocks.16.attn1.norm_k.weight": [ 1536 ], "blocks.16.attn2.to_q.bias": [ 1536 ], "blocks.16.attn2.to_q.packed_weight_indices": [ 1179648 ], "blocks.16.attn2.to_q.row_norms": [ 1536 ], "blocks.16.attn2.to_k.bias": [ 1536 ], "blocks.16.attn2.to_k.packed_weight_indices": [ 1179648 ], "blocks.16.attn2.to_k.row_norms": [ 1536 ], "blocks.16.attn2.to_v.bias": [ 1536 ], "blocks.16.attn2.to_v.packed_weight_indices": [ 1179648 ], "blocks.16.attn2.to_v.row_norms": [ 1536 ], "blocks.16.attn2.to_out.0.bias": [ 1536 ], "blocks.16.attn2.to_out.0.packed_weight_indices": [ 1179648 ], "blocks.16.attn2.to_out.0.row_norms": [ 1536 ], "blocks.16.attn2.norm_q.weight": [ 1536 ], "blocks.16.attn2.norm_k.weight": [ 1536 ], "blocks.16.norm2.weight": [ 1536 ], "blocks.16.norm2.bias": [ 1536 ], "blocks.16.ffn.net.0.proj.bias": [ 8960 ], "blocks.16.ffn.net.0.proj.packed_weight_indices": [ 6881280 ], "blocks.16.ffn.net.0.proj.row_norms": [ 8960 ], "blocks.16.ffn.net.2.bias": [ 1536 ], "blocks.16.ffn.net.2.packed_weight_indices": [ 6881280 ], "blocks.16.ffn.net.2.row_norms": [ 1536 ], "blocks.17.scale_shift_table": [ 1, 6, 1536 ], "blocks.17.attn1.to_q.bias": [ 1536 ], "blocks.17.attn1.to_q.packed_weight_indices": [ 1179648 ], "blocks.17.attn1.to_q.row_norms": [ 1536 ], "blocks.17.attn1.to_k.bias": [ 1536 ], "blocks.17.attn1.to_k.packed_weight_indices": [ 1179648 ], "blocks.17.attn1.to_k.row_norms": [ 1536 ], "blocks.17.attn1.to_v.bias": [ 1536 ], "blocks.17.attn1.to_v.packed_weight_indices": [ 1179648 ], "blocks.17.attn1.to_v.row_norms": [ 1536 ], "blocks.17.attn1.to_out.0.bias": [ 1536 ], "blocks.17.attn1.to_out.0.packed_weight_indices": [ 1179648 ], "blocks.17.attn1.to_out.0.row_norms": [ 1536 ], "blocks.17.attn1.norm_q.weight": [ 1536 ], "blocks.17.attn1.norm_k.weight": [ 1536 ], "blocks.17.attn2.to_q.bias": [ 1536 ], "blocks.17.attn2.to_q.packed_weight_indices": [ 1179648 ], "blocks.17.attn2.to_q.row_norms": [ 1536 ], "blocks.17.attn2.to_k.bias": [ 1536 ], "blocks.17.attn2.to_k.packed_weight_indices": [ 1179648 ], "blocks.17.attn2.to_k.row_norms": [ 1536 ], "blocks.17.attn2.to_v.bias": [ 1536 ], "blocks.17.attn2.to_v.packed_weight_indices": [ 1179648 ], "blocks.17.attn2.to_v.row_norms": [ 1536 ], "blocks.17.attn2.to_out.0.bias": [ 1536 ], "blocks.17.attn2.to_out.0.packed_weight_indices": [ 1179648 ], "blocks.17.attn2.to_out.0.row_norms": [ 1536 ], "blocks.17.attn2.norm_q.weight": [ 1536 ], "blocks.17.attn2.norm_k.weight": [ 1536 ], "blocks.17.norm2.weight": [ 1536 ], "blocks.17.norm2.bias": [ 1536 ], "blocks.17.ffn.net.0.proj.bias": [ 8960 ], "blocks.17.ffn.net.0.proj.packed_weight_indices": [ 6881280 ], "blocks.17.ffn.net.0.proj.row_norms": [ 8960 ], "blocks.17.ffn.net.2.bias": [ 1536 ], "blocks.17.ffn.net.2.packed_weight_indices": [ 6881280 ], "blocks.17.ffn.net.2.row_norms": [ 1536 ], "blocks.18.scale_shift_table": [ 1, 6, 1536 ], "blocks.18.attn1.to_q.bias": [ 1536 ], "blocks.18.attn1.to_q.packed_weight_indices": [ 1179648 ], "blocks.18.attn1.to_q.row_norms": [ 1536 ], "blocks.18.attn1.to_k.bias": [ 1536 ], "blocks.18.attn1.to_k.packed_weight_indices": [ 1179648 ], "blocks.18.attn1.to_k.row_norms": [ 1536 ], "blocks.18.attn1.to_v.bias": [ 1536 ], "blocks.18.attn1.to_v.packed_weight_indices": [ 1179648 ], "blocks.18.attn1.to_v.row_norms": [ 1536 ], "blocks.18.attn1.to_out.0.bias": [ 1536 ], "blocks.18.attn1.to_out.0.packed_weight_indices": [ 1179648 ], "blocks.18.attn1.to_out.0.row_norms": [ 1536 ], "blocks.18.attn1.norm_q.weight": [ 1536 ], "blocks.18.attn1.norm_k.weight": [ 1536 ], "blocks.18.attn2.to_q.bias": [ 1536 ], "blocks.18.attn2.to_q.packed_weight_indices": [ 1179648 ], "blocks.18.attn2.to_q.row_norms": [ 1536 ], "blocks.18.attn2.to_k.bias": [ 1536 ], "blocks.18.attn2.to_k.packed_weight_indices": [ 1179648 ], "blocks.18.attn2.to_k.row_norms": [ 1536 ], "blocks.18.attn2.to_v.bias": [ 1536 ], "blocks.18.attn2.to_v.packed_weight_indices": [ 1179648 ], "blocks.18.attn2.to_v.row_norms": [ 1536 ], "blocks.18.attn2.to_out.0.bias": [ 1536 ], "blocks.18.attn2.to_out.0.packed_weight_indices": [ 1179648 ], "blocks.18.attn2.to_out.0.row_norms": [ 1536 ], "blocks.18.attn2.norm_q.weight": [ 1536 ], "blocks.18.attn2.norm_k.weight": [ 1536 ], "blocks.18.norm2.weight": [ 1536 ], "blocks.18.norm2.bias": [ 1536 ], "blocks.18.ffn.net.0.proj.bias": [ 8960 ], "blocks.18.ffn.net.0.proj.packed_weight_indices": [ 6881280 ], "blocks.18.ffn.net.0.proj.row_norms": [ 8960 ], "blocks.18.ffn.net.2.bias": [ 1536 ], "blocks.18.ffn.net.2.packed_weight_indices": [ 6881280 ], "blocks.18.ffn.net.2.row_norms": [ 1536 ], "blocks.19.scale_shift_table": [ 1, 6, 1536 ], "blocks.19.attn1.to_q.bias": [ 1536 ], "blocks.19.attn1.to_q.packed_weight_indices": [ 1179648 ], "blocks.19.attn1.to_q.row_norms": [ 1536 ], "blocks.19.attn1.to_k.bias": [ 1536 ], "blocks.19.attn1.to_k.packed_weight_indices": [ 1179648 ], "blocks.19.attn1.to_k.row_norms": [ 1536 ], "blocks.19.attn1.to_v.bias": [ 1536 ], "blocks.19.attn1.to_v.packed_weight_indices": [ 1179648 ], "blocks.19.attn1.to_v.row_norms": [ 1536 ], "blocks.19.attn1.to_out.0.bias": [ 1536 ], "blocks.19.attn1.to_out.0.packed_weight_indices": [ 1179648 ], "blocks.19.attn1.to_out.0.row_norms": [ 1536 ], "blocks.19.attn1.norm_q.weight": [ 1536 ], "blocks.19.attn1.norm_k.weight": [ 1536 ], "blocks.19.attn2.to_q.bias": [ 1536 ], "blocks.19.attn2.to_q.packed_weight_indices": [ 1179648 ], "blocks.19.attn2.to_q.row_norms": [ 1536 ], "blocks.19.attn2.to_k.bias": [ 1536 ], "blocks.19.attn2.to_k.packed_weight_indices": [ 1179648 ], "blocks.19.attn2.to_k.row_norms": [ 1536 ], "blocks.19.attn2.to_v.bias": [ 1536 ], "blocks.19.attn2.to_v.packed_weight_indices": [ 1179648 ], "blocks.19.attn2.to_v.row_norms": [ 1536 ], "blocks.19.attn2.to_out.0.bias": [ 1536 ], "blocks.19.attn2.to_out.0.packed_weight_indices": [ 1179648 ], "blocks.19.attn2.to_out.0.row_norms": [ 1536 ], "blocks.19.attn2.norm_q.weight": [ 1536 ], "blocks.19.attn2.norm_k.weight": [ 1536 ], "blocks.19.norm2.weight": [ 1536 ], "blocks.19.norm2.bias": [ 1536 ], "blocks.19.ffn.net.0.proj.bias": [ 8960 ], "blocks.19.ffn.net.0.proj.packed_weight_indices": [ 6881280 ], "blocks.19.ffn.net.0.proj.row_norms": [ 8960 ], "blocks.19.ffn.net.2.bias": [ 1536 ], "blocks.19.ffn.net.2.packed_weight_indices": [ 6881280 ], "blocks.19.ffn.net.2.row_norms": [ 1536 ], "blocks.20.scale_shift_table": [ 1, 6, 1536 ], "blocks.20.attn1.to_q.bias": [ 1536 ], "blocks.20.attn1.to_q.packed_weight_indices": [ 1179648 ], "blocks.20.attn1.to_q.row_norms": [ 1536 ], "blocks.20.attn1.to_k.bias": [ 1536 ], "blocks.20.attn1.to_k.packed_weight_indices": [ 1179648 ], "blocks.20.attn1.to_k.row_norms": [ 1536 ], "blocks.20.attn1.to_v.bias": [ 1536 ], "blocks.20.attn1.to_v.packed_weight_indices": [ 1179648 ], "blocks.20.attn1.to_v.row_norms": [ 1536 ], "blocks.20.attn1.to_out.0.bias": [ 1536 ], "blocks.20.attn1.to_out.0.packed_weight_indices": [ 1179648 ], "blocks.20.attn1.to_out.0.row_norms": [ 1536 ], "blocks.20.attn1.norm_q.weight": [ 1536 ], "blocks.20.attn1.norm_k.weight": [ 1536 ], "blocks.20.attn2.to_q.bias": [ 1536 ], "blocks.20.attn2.to_q.packed_weight_indices": [ 1179648 ], "blocks.20.attn2.to_q.row_norms": [ 1536 ], "blocks.20.attn2.to_k.bias": [ 1536 ], "blocks.20.attn2.to_k.packed_weight_indices": [ 1179648 ], "blocks.20.attn2.to_k.row_norms": [ 1536 ], "blocks.20.attn2.to_v.bias": [ 1536 ], "blocks.20.attn2.to_v.packed_weight_indices": [ 1179648 ], "blocks.20.attn2.to_v.row_norms": [ 1536 ], "blocks.20.attn2.to_out.0.bias": [ 1536 ], "blocks.20.attn2.to_out.0.packed_weight_indices": [ 1179648 ], "blocks.20.attn2.to_out.0.row_norms": [ 1536 ], "blocks.20.attn2.norm_q.weight": [ 1536 ], "blocks.20.attn2.norm_k.weight": [ 1536 ], "blocks.20.norm2.weight": [ 1536 ], "blocks.20.norm2.bias": [ 1536 ], "blocks.20.ffn.net.0.proj.bias": [ 8960 ], "blocks.20.ffn.net.0.proj.packed_weight_indices": [ 6881280 ], "blocks.20.ffn.net.0.proj.row_norms": [ 8960 ], "blocks.20.ffn.net.2.bias": [ 1536 ], "blocks.20.ffn.net.2.packed_weight_indices": [ 6881280 ], "blocks.20.ffn.net.2.row_norms": [ 1536 ], "blocks.21.scale_shift_table": [ 1, 6, 1536 ], "blocks.21.attn1.to_q.bias": [ 1536 ], "blocks.21.attn1.to_q.packed_weight_indices": [ 1179648 ], "blocks.21.attn1.to_q.row_norms": [ 1536 ], "blocks.21.attn1.to_k.bias": [ 1536 ], "blocks.21.attn1.to_k.packed_weight_indices": [ 1179648 ], "blocks.21.attn1.to_k.row_norms": [ 1536 ], "blocks.21.attn1.to_v.bias": [ 1536 ], "blocks.21.attn1.to_v.packed_weight_indices": [ 1179648 ], "blocks.21.attn1.to_v.row_norms": [ 1536 ], "blocks.21.attn1.to_out.0.bias": [ 1536 ], "blocks.21.attn1.to_out.0.packed_weight_indices": [ 1179648 ], "blocks.21.attn1.to_out.0.row_norms": [ 1536 ], "blocks.21.attn1.norm_q.weight": [ 1536 ], "blocks.21.attn1.norm_k.weight": [ 1536 ], "blocks.21.attn2.to_q.bias": [ 1536 ], "blocks.21.attn2.to_q.packed_weight_indices": [ 1179648 ], "blocks.21.attn2.to_q.row_norms": [ 1536 ], "blocks.21.attn2.to_k.bias": [ 1536 ], "blocks.21.attn2.to_k.packed_weight_indices": [ 1179648 ], "blocks.21.attn2.to_k.row_norms": [ 1536 ], "blocks.21.attn2.to_v.bias": [ 1536 ], "blocks.21.attn2.to_v.packed_weight_indices": [ 1179648 ], "blocks.21.attn2.to_v.row_norms": [ 1536 ], "blocks.21.attn2.to_out.0.bias": [ 1536 ], "blocks.21.attn2.to_out.0.packed_weight_indices": [ 1179648 ], "blocks.21.attn2.to_out.0.row_norms": [ 1536 ], "blocks.21.attn2.norm_q.weight": [ 1536 ], "blocks.21.attn2.norm_k.weight": [ 1536 ], "blocks.21.norm2.weight": [ 1536 ], "blocks.21.norm2.bias": [ 1536 ], "blocks.21.ffn.net.0.proj.bias": [ 8960 ], "blocks.21.ffn.net.0.proj.packed_weight_indices": [ 6881280 ], "blocks.21.ffn.net.0.proj.row_norms": [ 8960 ], "blocks.21.ffn.net.2.bias": [ 1536 ], "blocks.21.ffn.net.2.packed_weight_indices": [ 6881280 ], "blocks.21.ffn.net.2.row_norms": [ 1536 ], "blocks.22.scale_shift_table": [ 1, 6, 1536 ], "blocks.22.attn1.to_q.bias": [ 1536 ], "blocks.22.attn1.to_q.packed_weight_indices": [ 1179648 ], "blocks.22.attn1.to_q.row_norms": [ 1536 ], "blocks.22.attn1.to_k.bias": [ 1536 ], "blocks.22.attn1.to_k.packed_weight_indices": [ 1179648 ], "blocks.22.attn1.to_k.row_norms": [ 1536 ], "blocks.22.attn1.to_v.bias": [ 1536 ], "blocks.22.attn1.to_v.packed_weight_indices": [ 1179648 ], "blocks.22.attn1.to_v.row_norms": [ 1536 ], "blocks.22.attn1.to_out.0.bias": [ 1536 ], "blocks.22.attn1.to_out.0.packed_weight_indices": [ 1179648 ], "blocks.22.attn1.to_out.0.row_norms": [ 1536 ], "blocks.22.attn1.norm_q.weight": [ 1536 ], "blocks.22.attn1.norm_k.weight": [ 1536 ], "blocks.22.attn2.to_q.bias": [ 1536 ], "blocks.22.attn2.to_q.packed_weight_indices": [ 1179648 ], "blocks.22.attn2.to_q.row_norms": [ 1536 ], "blocks.22.attn2.to_k.bias": [ 1536 ], "blocks.22.attn2.to_k.packed_weight_indices": [ 1179648 ], "blocks.22.attn2.to_k.row_norms": [ 1536 ], "blocks.22.attn2.to_v.bias": [ 1536 ], "blocks.22.attn2.to_v.packed_weight_indices": [ 1179648 ], "blocks.22.attn2.to_v.row_norms": [ 1536 ], "blocks.22.attn2.to_out.0.bias": [ 1536 ], "blocks.22.attn2.to_out.0.packed_weight_indices": [ 1179648 ], "blocks.22.attn2.to_out.0.row_norms": [ 1536 ], "blocks.22.attn2.norm_q.weight": [ 1536 ], "blocks.22.attn2.norm_k.weight": [ 1536 ], "blocks.22.norm2.weight": [ 1536 ], "blocks.22.norm2.bias": [ 1536 ], "blocks.22.ffn.net.0.proj.bias": [ 8960 ], "blocks.22.ffn.net.0.proj.packed_weight_indices": [ 6881280 ], "blocks.22.ffn.net.0.proj.row_norms": [ 8960 ], "blocks.22.ffn.net.2.bias": [ 1536 ], "blocks.22.ffn.net.2.packed_weight_indices": [ 6881280 ], "blocks.22.ffn.net.2.row_norms": [ 1536 ], "blocks.23.scale_shift_table": [ 1, 6, 1536 ], "blocks.23.attn1.to_q.bias": [ 1536 ], "blocks.23.attn1.to_q.packed_weight_indices": [ 1179648 ], "blocks.23.attn1.to_q.row_norms": [ 1536 ], "blocks.23.attn1.to_k.bias": [ 1536 ], "blocks.23.attn1.to_k.packed_weight_indices": [ 1179648 ], "blocks.23.attn1.to_k.row_norms": [ 1536 ], "blocks.23.attn1.to_v.bias": [ 1536 ], "blocks.23.attn1.to_v.packed_weight_indices": [ 1179648 ], "blocks.23.attn1.to_v.row_norms": [ 1536 ], "blocks.23.attn1.to_out.0.bias": [ 1536 ], "blocks.23.attn1.to_out.0.packed_weight_indices": [ 1179648 ], "blocks.23.attn1.to_out.0.row_norms": [ 1536 ], "blocks.23.attn1.norm_q.weight": [ 1536 ], "blocks.23.attn1.norm_k.weight": [ 1536 ], "blocks.23.attn2.to_q.bias": [ 1536 ], "blocks.23.attn2.to_q.packed_weight_indices": [ 1179648 ], "blocks.23.attn2.to_q.row_norms": [ 1536 ], "blocks.23.attn2.to_k.bias": [ 1536 ], "blocks.23.attn2.to_k.packed_weight_indices": [ 1179648 ], "blocks.23.attn2.to_k.row_norms": [ 1536 ], "blocks.23.attn2.to_v.bias": [ 1536 ], "blocks.23.attn2.to_v.packed_weight_indices": [ 1179648 ], "blocks.23.attn2.to_v.row_norms": [ 1536 ], "blocks.23.attn2.to_out.0.bias": [ 1536 ], "blocks.23.attn2.to_out.0.packed_weight_indices": [ 1179648 ], "blocks.23.attn2.to_out.0.row_norms": [ 1536 ], "blocks.23.attn2.norm_q.weight": [ 1536 ], "blocks.23.attn2.norm_k.weight": [ 1536 ], "blocks.23.norm2.weight": [ 1536 ], "blocks.23.norm2.bias": [ 1536 ], "blocks.23.ffn.net.0.proj.bias": [ 8960 ], "blocks.23.ffn.net.0.proj.packed_weight_indices": [ 6881280 ], "blocks.23.ffn.net.0.proj.row_norms": [ 8960 ], "blocks.23.ffn.net.2.bias": [ 1536 ], "blocks.23.ffn.net.2.packed_weight_indices": [ 6881280 ], "blocks.23.ffn.net.2.row_norms": [ 1536 ], "blocks.24.scale_shift_table": [ 1, 6, 1536 ], "blocks.24.attn1.to_q.bias": [ 1536 ], "blocks.24.attn1.to_q.packed_weight_indices": [ 1179648 ], "blocks.24.attn1.to_q.row_norms": [ 1536 ], "blocks.24.attn1.to_k.bias": [ 1536 ], "blocks.24.attn1.to_k.packed_weight_indices": [ 1179648 ], "blocks.24.attn1.to_k.row_norms": [ 1536 ], "blocks.24.attn1.to_v.bias": [ 1536 ], "blocks.24.attn1.to_v.packed_weight_indices": [ 1179648 ], "blocks.24.attn1.to_v.row_norms": [ 1536 ], "blocks.24.attn1.to_out.0.bias": [ 1536 ], "blocks.24.attn1.to_out.0.packed_weight_indices": [ 1179648 ], "blocks.24.attn1.to_out.0.row_norms": [ 1536 ], "blocks.24.attn1.norm_q.weight": [ 1536 ], "blocks.24.attn1.norm_k.weight": [ 1536 ], "blocks.24.attn2.to_q.bias": [ 1536 ], "blocks.24.attn2.to_q.packed_weight_indices": [ 1179648 ], "blocks.24.attn2.to_q.row_norms": [ 1536 ], "blocks.24.attn2.to_k.bias": [ 1536 ], "blocks.24.attn2.to_k.packed_weight_indices": [ 1179648 ], "blocks.24.attn2.to_k.row_norms": [ 1536 ], "blocks.24.attn2.to_v.bias": [ 1536 ], "blocks.24.attn2.to_v.packed_weight_indices": [ 1179648 ], "blocks.24.attn2.to_v.row_norms": [ 1536 ], "blocks.24.attn2.to_out.0.bias": [ 1536 ], "blocks.24.attn2.to_out.0.packed_weight_indices": [ 1179648 ], "blocks.24.attn2.to_out.0.row_norms": [ 1536 ], "blocks.24.attn2.norm_q.weight": [ 1536 ], "blocks.24.attn2.norm_k.weight": [ 1536 ], "blocks.24.norm2.weight": [ 1536 ], "blocks.24.norm2.bias": [ 1536 ], "blocks.24.ffn.net.0.proj.bias": [ 8960 ], "blocks.24.ffn.net.0.proj.packed_weight_indices": [ 6881280 ], "blocks.24.ffn.net.0.proj.row_norms": [ 8960 ], "blocks.24.ffn.net.2.bias": [ 1536 ], "blocks.24.ffn.net.2.packed_weight_indices": [ 6881280 ], "blocks.24.ffn.net.2.row_norms": [ 1536 ], "blocks.25.scale_shift_table": [ 1, 6, 1536 ], "blocks.25.attn1.to_q.bias": [ 1536 ], "blocks.25.attn1.to_q.packed_weight_indices": [ 1179648 ], "blocks.25.attn1.to_q.row_norms": [ 1536 ], "blocks.25.attn1.to_k.bias": [ 1536 ], "blocks.25.attn1.to_k.packed_weight_indices": [ 1179648 ], "blocks.25.attn1.to_k.row_norms": [ 1536 ], "blocks.25.attn1.to_v.bias": [ 1536 ], "blocks.25.attn1.to_v.packed_weight_indices": [ 1179648 ], "blocks.25.attn1.to_v.row_norms": [ 1536 ], "blocks.25.attn1.to_out.0.bias": [ 1536 ], "blocks.25.attn1.to_out.0.packed_weight_indices": [ 1179648 ], "blocks.25.attn1.to_out.0.row_norms": [ 1536 ], "blocks.25.attn1.norm_q.weight": [ 1536 ], "blocks.25.attn1.norm_k.weight": [ 1536 ], "blocks.25.attn2.to_q.bias": [ 1536 ], "blocks.25.attn2.to_q.packed_weight_indices": [ 1179648 ], "blocks.25.attn2.to_q.row_norms": [ 1536 ], "blocks.25.attn2.to_k.bias": [ 1536 ], "blocks.25.attn2.to_k.packed_weight_indices": [ 1179648 ], "blocks.25.attn2.to_k.row_norms": [ 1536 ], "blocks.25.attn2.to_v.bias": [ 1536 ], "blocks.25.attn2.to_v.packed_weight_indices": [ 1179648 ], "blocks.25.attn2.to_v.row_norms": [ 1536 ], "blocks.25.attn2.to_out.0.bias": [ 1536 ], "blocks.25.attn2.to_out.0.packed_weight_indices": [ 1179648 ], "blocks.25.attn2.to_out.0.row_norms": [ 1536 ], "blocks.25.attn2.norm_q.weight": [ 1536 ], "blocks.25.attn2.norm_k.weight": [ 1536 ], "blocks.25.norm2.weight": [ 1536 ], "blocks.25.norm2.bias": [ 1536 ], "blocks.25.ffn.net.0.proj.bias": [ 8960 ], "blocks.25.ffn.net.0.proj.packed_weight_indices": [ 6881280 ], "blocks.25.ffn.net.0.proj.row_norms": [ 8960 ], "blocks.25.ffn.net.2.bias": [ 1536 ], "blocks.25.ffn.net.2.packed_weight_indices": [ 6881280 ], "blocks.25.ffn.net.2.row_norms": [ 1536 ], "blocks.26.scale_shift_table": [ 1, 6, 1536 ], "blocks.26.attn1.to_q.bias": [ 1536 ], "blocks.26.attn1.to_q.packed_weight_indices": [ 1179648 ], "blocks.26.attn1.to_q.row_norms": [ 1536 ], "blocks.26.attn1.to_k.bias": [ 1536 ], "blocks.26.attn1.to_k.packed_weight_indices": [ 1179648 ], "blocks.26.attn1.to_k.row_norms": [ 1536 ], "blocks.26.attn1.to_v.bias": [ 1536 ], "blocks.26.attn1.to_v.packed_weight_indices": [ 1179648 ], "blocks.26.attn1.to_v.row_norms": [ 1536 ], "blocks.26.attn1.to_out.0.bias": [ 1536 ], "blocks.26.attn1.to_out.0.packed_weight_indices": [ 1179648 ], "blocks.26.attn1.to_out.0.row_norms": [ 1536 ], "blocks.26.attn1.norm_q.weight": [ 1536 ], "blocks.26.attn1.norm_k.weight": [ 1536 ], "blocks.26.attn2.to_q.bias": [ 1536 ], "blocks.26.attn2.to_q.packed_weight_indices": [ 1179648 ], "blocks.26.attn2.to_q.row_norms": [ 1536 ], "blocks.26.attn2.to_k.bias": [ 1536 ], "blocks.26.attn2.to_k.packed_weight_indices": [ 1179648 ], "blocks.26.attn2.to_k.row_norms": [ 1536 ], "blocks.26.attn2.to_v.bias": [ 1536 ], "blocks.26.attn2.to_v.packed_weight_indices": [ 1179648 ], "blocks.26.attn2.to_v.row_norms": [ 1536 ], "blocks.26.attn2.to_out.0.bias": [ 1536 ], "blocks.26.attn2.to_out.0.packed_weight_indices": [ 1179648 ], "blocks.26.attn2.to_out.0.row_norms": [ 1536 ], "blocks.26.attn2.norm_q.weight": [ 1536 ], "blocks.26.attn2.norm_k.weight": [ 1536 ], "blocks.26.norm2.weight": [ 1536 ], "blocks.26.norm2.bias": [ 1536 ], "blocks.26.ffn.net.0.proj.bias": [ 8960 ], "blocks.26.ffn.net.0.proj.packed_weight_indices": [ 6881280 ], "blocks.26.ffn.net.0.proj.row_norms": [ 8960 ], "blocks.26.ffn.net.2.bias": [ 1536 ], "blocks.26.ffn.net.2.packed_weight_indices": [ 6881280 ], "blocks.26.ffn.net.2.row_norms": [ 1536 ], "blocks.27.scale_shift_table": [ 1, 6, 1536 ], "blocks.27.attn1.to_q.bias": [ 1536 ], "blocks.27.attn1.to_q.packed_weight_indices": [ 1179648 ], "blocks.27.attn1.to_q.row_norms": [ 1536 ], "blocks.27.attn1.to_k.bias": [ 1536 ], "blocks.27.attn1.to_k.packed_weight_indices": [ 1179648 ], "blocks.27.attn1.to_k.row_norms": [ 1536 ], "blocks.27.attn1.to_v.bias": [ 1536 ], "blocks.27.attn1.to_v.packed_weight_indices": [ 1179648 ], "blocks.27.attn1.to_v.row_norms": [ 1536 ], "blocks.27.attn1.to_out.0.bias": [ 1536 ], "blocks.27.attn1.to_out.0.packed_weight_indices": [ 1179648 ], "blocks.27.attn1.to_out.0.row_norms": [ 1536 ], "blocks.27.attn1.norm_q.weight": [ 1536 ], "blocks.27.attn1.norm_k.weight": [ 1536 ], "blocks.27.attn2.to_q.bias": [ 1536 ], "blocks.27.attn2.to_q.packed_weight_indices": [ 1179648 ], "blocks.27.attn2.to_q.row_norms": [ 1536 ], "blocks.27.attn2.to_k.bias": [ 1536 ], "blocks.27.attn2.to_k.packed_weight_indices": [ 1179648 ], "blocks.27.attn2.to_k.row_norms": [ 1536 ], "blocks.27.attn2.to_v.bias": [ 1536 ], "blocks.27.attn2.to_v.packed_weight_indices": [ 1179648 ], "blocks.27.attn2.to_v.row_norms": [ 1536 ], "blocks.27.attn2.to_out.0.bias": [ 1536 ], "blocks.27.attn2.to_out.0.packed_weight_indices": [ 1179648 ], "blocks.27.attn2.to_out.0.row_norms": [ 1536 ], "blocks.27.attn2.norm_q.weight": [ 1536 ], "blocks.27.attn2.norm_k.weight": [ 1536 ], "blocks.27.norm2.weight": [ 1536 ], "blocks.27.norm2.bias": [ 1536 ], "blocks.27.ffn.net.0.proj.bias": [ 8960 ], "blocks.27.ffn.net.0.proj.packed_weight_indices": [ 6881280 ], "blocks.27.ffn.net.0.proj.row_norms": [ 8960 ], "blocks.27.ffn.net.2.bias": [ 1536 ], "blocks.27.ffn.net.2.packed_weight_indices": [ 6881280 ], "blocks.27.ffn.net.2.row_norms": [ 1536 ], "blocks.28.scale_shift_table": [ 1, 6, 1536 ], "blocks.28.attn1.to_q.bias": [ 1536 ], "blocks.28.attn1.to_q.packed_weight_indices": [ 1179648 ], "blocks.28.attn1.to_q.row_norms": [ 1536 ], "blocks.28.attn1.to_k.bias": [ 1536 ], "blocks.28.attn1.to_k.packed_weight_indices": [ 1179648 ], "blocks.28.attn1.to_k.row_norms": [ 1536 ], "blocks.28.attn1.to_v.bias": [ 1536 ], "blocks.28.attn1.to_v.packed_weight_indices": [ 1179648 ], "blocks.28.attn1.to_v.row_norms": [ 1536 ], "blocks.28.attn1.to_out.0.bias": [ 1536 ], "blocks.28.attn1.to_out.0.packed_weight_indices": [ 1179648 ], "blocks.28.attn1.to_out.0.row_norms": [ 1536 ], "blocks.28.attn1.norm_q.weight": [ 1536 ], "blocks.28.attn1.norm_k.weight": [ 1536 ], "blocks.28.attn2.to_q.bias": [ 1536 ], "blocks.28.attn2.to_q.packed_weight_indices": [ 1179648 ], "blocks.28.attn2.to_q.row_norms": [ 1536 ], "blocks.28.attn2.to_k.bias": [ 1536 ], "blocks.28.attn2.to_k.packed_weight_indices": [ 1179648 ], "blocks.28.attn2.to_k.row_norms": [ 1536 ], "blocks.28.attn2.to_v.bias": [ 1536 ], "blocks.28.attn2.to_v.packed_weight_indices": [ 1179648 ], "blocks.28.attn2.to_v.row_norms": [ 1536 ], "blocks.28.attn2.to_out.0.bias": [ 1536 ], "blocks.28.attn2.to_out.0.packed_weight_indices": [ 1179648 ], "blocks.28.attn2.to_out.0.row_norms": [ 1536 ], "blocks.28.attn2.norm_q.weight": [ 1536 ], "blocks.28.attn2.norm_k.weight": [ 1536 ], "blocks.28.norm2.weight": [ 1536 ], "blocks.28.norm2.bias": [ 1536 ], "blocks.28.ffn.net.0.proj.bias": [ 8960 ], "blocks.28.ffn.net.0.proj.packed_weight_indices": [ 6881280 ], "blocks.28.ffn.net.0.proj.row_norms": [ 8960 ], "blocks.28.ffn.net.2.bias": [ 1536 ], "blocks.28.ffn.net.2.packed_weight_indices": [ 6881280 ], "blocks.28.ffn.net.2.row_norms": [ 1536 ], "blocks.29.scale_shift_table": [ 1, 6, 1536 ], "blocks.29.attn1.to_q.bias": [ 1536 ], "blocks.29.attn1.to_q.packed_weight_indices": [ 1179648 ], "blocks.29.attn1.to_q.row_norms": [ 1536 ], "blocks.29.attn1.to_k.bias": [ 1536 ], "blocks.29.attn1.to_k.packed_weight_indices": [ 1179648 ], "blocks.29.attn1.to_k.row_norms": [ 1536 ], "blocks.29.attn1.to_v.bias": [ 1536 ], "blocks.29.attn1.to_v.packed_weight_indices": [ 1179648 ], "blocks.29.attn1.to_v.row_norms": [ 1536 ], "blocks.29.attn1.to_out.0.bias": [ 1536 ], "blocks.29.attn1.to_out.0.packed_weight_indices": [ 1179648 ], "blocks.29.attn1.to_out.0.row_norms": [ 1536 ], "blocks.29.attn1.norm_q.weight": [ 1536 ], "blocks.29.attn1.norm_k.weight": [ 1536 ], "blocks.29.attn2.to_q.bias": [ 1536 ], "blocks.29.attn2.to_q.packed_weight_indices": [ 1179648 ], "blocks.29.attn2.to_q.row_norms": [ 1536 ], "blocks.29.attn2.to_k.bias": [ 1536 ], "blocks.29.attn2.to_k.packed_weight_indices": [ 1179648 ], "blocks.29.attn2.to_k.row_norms": [ 1536 ], "blocks.29.attn2.to_v.bias": [ 1536 ], "blocks.29.attn2.to_v.packed_weight_indices": [ 1179648 ], "blocks.29.attn2.to_v.row_norms": [ 1536 ], "blocks.29.attn2.to_out.0.bias": [ 1536 ], "blocks.29.attn2.to_out.0.packed_weight_indices": [ 1179648 ], "blocks.29.attn2.to_out.0.row_norms": [ 1536 ], "blocks.29.attn2.norm_q.weight": [ 1536 ], "blocks.29.attn2.norm_k.weight": [ 1536 ], "blocks.29.norm2.weight": [ 1536 ], "blocks.29.norm2.bias": [ 1536 ], "blocks.29.ffn.net.0.proj.bias": [ 8960 ], "blocks.29.ffn.net.0.proj.packed_weight_indices": [ 6881280 ], "blocks.29.ffn.net.0.proj.row_norms": [ 8960 ], "blocks.29.ffn.net.2.bias": [ 1536 ], "blocks.29.ffn.net.2.packed_weight_indices": [ 6881280 ], "blocks.29.ffn.net.2.row_norms": [ 1536 ], "proj_out.weight": [ 64, 1536 ], "proj_out.bias": [ 64 ] }, "checksums": { "assets/video_generation_comparison_matrix.webp": "de523743fa4c42852524cd7635d6ace1d11149305aa8615ffd788a4bf59aeb46", "benchmark/summary.json": "14966a818ef2b61424bf02154a0414bdad5c4583d444dcbafaf87b056cde6dfe", "model.safetensors": "5778da463b35c5c68761c7307e05634f3c818e2a05d481eeb50799cceb376889", "model_index.json": "923430fc4a686c5d0a5f1a82d5b3b40eeeececfc3773eb26a31cf4668b3f70a3", "orbitquant_codebooks.safetensors": "2b534248778a59626b9c21e32ccf4760d4061a06b6aa7fcfdacb5a37a660605e", "orbitquant_rotations.safetensors": "b7b2d1e28dd66554c2539d9152b63938a86f1fae0d65664a0df786313dee5dd9", "prompts.json": "b01eb3abccf0db3cb1699c09033db918ceb7886cd9b6b6d4f03304ed330ddbc8", "quantization_config.json": "f8a5a7dc5e808adade3f064fa556d535d4bdc5c13a778552929347939f2fc97c" } }