| { | |
| "dfloat11_config": { | |
| "bytes_per_thread": 8, | |
| "pattern_dict": { | |
| "language_model\\.model\\.embed_tokens": [], | |
| "language_model\\.model\\.layers.\\d+": [ | |
| "self_attn.q_proj", | |
| "self_attn.k_proj", | |
| "self_attn.v_proj", | |
| "self_attn.o_proj", | |
| "self_attn.q_proj_moe_gen", | |
| "self_attn.k_proj_moe_gen", | |
| "self_attn.v_proj_moe_gen", | |
| "self_attn.o_proj_moe_gen", | |
| "mlp.gate_proj", | |
| "mlp.up_proj", | |
| "mlp.down_proj", | |
| "mlp_moe_gen.gate_proj", | |
| "mlp_moe_gen.up_proj", | |
| "mlp_moe_gen.down_proj" | |
| ], | |
| "language_model\\.lm_head": [], | |
| "vit_model": [ | |
| "vision_model.encoder.layers.0.self_attn.k_proj", | |
| "vision_model.encoder.layers.0.self_attn.v_proj", | |
| "vision_model.encoder.layers.0.self_attn.q_proj", | |
| "vision_model.encoder.layers.0.self_attn.out_proj", | |
| "vision_model.encoder.layers.0.mlp.fc1", | |
| "vision_model.encoder.layers.0.mlp.fc2", | |
| "vision_model.encoder.layers.1.self_attn.k_proj", | |
| "vision_model.encoder.layers.1.self_attn.v_proj", | |
| "vision_model.encoder.layers.1.self_attn.q_proj", | |
| "vision_model.encoder.layers.1.self_attn.out_proj", | |
| "vision_model.encoder.layers.1.mlp.fc1", | |
| "vision_model.encoder.layers.1.mlp.fc2", | |
| "vision_model.encoder.layers.2.self_attn.k_proj", | |
| "vision_model.encoder.layers.2.self_attn.v_proj", | |
| "vision_model.encoder.layers.2.self_attn.q_proj", | |
| "vision_model.encoder.layers.2.self_attn.out_proj", | |
| "vision_model.encoder.layers.2.mlp.fc1", | |
| "vision_model.encoder.layers.2.mlp.fc2", | |
| "vision_model.encoder.layers.3.self_attn.k_proj", | |
| "vision_model.encoder.layers.3.self_attn.v_proj", | |
| "vision_model.encoder.layers.3.self_attn.q_proj", | |
| "vision_model.encoder.layers.3.self_attn.out_proj", | |
| "vision_model.encoder.layers.3.mlp.fc1", | |
| "vision_model.encoder.layers.3.mlp.fc2", | |
| "vision_model.encoder.layers.4.self_attn.k_proj", | |
| "vision_model.encoder.layers.4.self_attn.v_proj", | |
| "vision_model.encoder.layers.4.self_attn.q_proj", | |
| "vision_model.encoder.layers.4.self_attn.out_proj", | |
| "vision_model.encoder.layers.4.mlp.fc1", | |
| "vision_model.encoder.layers.4.mlp.fc2", | |
| "vision_model.encoder.layers.5.self_attn.k_proj", | |
| "vision_model.encoder.layers.5.self_attn.v_proj", | |
| "vision_model.encoder.layers.5.self_attn.q_proj", | |
| "vision_model.encoder.layers.5.self_attn.out_proj", | |
| "vision_model.encoder.layers.5.mlp.fc1", | |
| "vision_model.encoder.layers.5.mlp.fc2", | |
| "vision_model.encoder.layers.6.self_attn.k_proj", | |
| "vision_model.encoder.layers.6.self_attn.v_proj", | |
| "vision_model.encoder.layers.6.self_attn.q_proj", | |
| "vision_model.encoder.layers.6.self_attn.out_proj", | |
| "vision_model.encoder.layers.6.mlp.fc1", | |
| "vision_model.encoder.layers.6.mlp.fc2", | |
| "vision_model.encoder.layers.7.self_attn.k_proj", | |
| "vision_model.encoder.layers.7.self_attn.v_proj", | |
| "vision_model.encoder.layers.7.self_attn.q_proj", | |
| "vision_model.encoder.layers.7.self_attn.out_proj", | |
| "vision_model.encoder.layers.7.mlp.fc1", | |
| "vision_model.encoder.layers.7.mlp.fc2", | |
| "vision_model.encoder.layers.8.self_attn.k_proj", | |
| "vision_model.encoder.layers.8.self_attn.v_proj", | |
| "vision_model.encoder.layers.8.self_attn.q_proj", | |
| "vision_model.encoder.layers.8.self_attn.out_proj", | |
| "vision_model.encoder.layers.8.mlp.fc1", | |
| "vision_model.encoder.layers.8.mlp.fc2", | |
| "vision_model.encoder.layers.9.self_attn.k_proj", | |
| "vision_model.encoder.layers.9.self_attn.v_proj", | |
| "vision_model.encoder.layers.9.self_attn.q_proj", | |
| "vision_model.encoder.layers.9.self_attn.out_proj", | |
| "vision_model.encoder.layers.9.mlp.fc1", | |
| "vision_model.encoder.layers.9.mlp.fc2", | |
| "vision_model.encoder.layers.10.self_attn.k_proj", | |
| "vision_model.encoder.layers.10.self_attn.v_proj", | |
| "vision_model.encoder.layers.10.self_attn.q_proj", | |
| "vision_model.encoder.layers.10.self_attn.out_proj", | |
| "vision_model.encoder.layers.10.mlp.fc1", | |
| "vision_model.encoder.layers.10.mlp.fc2", | |
| "vision_model.encoder.layers.11.self_attn.k_proj", | |
| "vision_model.encoder.layers.11.self_attn.v_proj", | |
| "vision_model.encoder.layers.11.self_attn.q_proj", | |
| "vision_model.encoder.layers.11.self_attn.out_proj", | |
| "vision_model.encoder.layers.11.mlp.fc1", | |
| "vision_model.encoder.layers.11.mlp.fc2", | |
| "vision_model.encoder.layers.12.self_attn.k_proj", | |
| "vision_model.encoder.layers.12.self_attn.v_proj", | |
| "vision_model.encoder.layers.12.self_attn.q_proj", | |
| "vision_model.encoder.layers.12.self_attn.out_proj", | |
| "vision_model.encoder.layers.12.mlp.fc1", | |
| "vision_model.encoder.layers.12.mlp.fc2", | |
| "vision_model.encoder.layers.13.self_attn.k_proj", | |
| "vision_model.encoder.layers.13.self_attn.v_proj", | |
| "vision_model.encoder.layers.13.self_attn.q_proj", | |
| "vision_model.encoder.layers.13.self_attn.out_proj", | |
| "vision_model.encoder.layers.13.mlp.fc1", | |
| "vision_model.encoder.layers.13.mlp.fc2", | |
| "vision_model.encoder.layers.14.self_attn.k_proj", | |
| "vision_model.encoder.layers.14.self_attn.v_proj", | |
| "vision_model.encoder.layers.14.self_attn.q_proj", | |
| "vision_model.encoder.layers.14.self_attn.out_proj", | |
| "vision_model.encoder.layers.14.mlp.fc1", | |
| "vision_model.encoder.layers.14.mlp.fc2", | |
| "vision_model.encoder.layers.15.self_attn.k_proj", | |
| "vision_model.encoder.layers.15.self_attn.v_proj", | |
| "vision_model.encoder.layers.15.self_attn.q_proj", | |
| "vision_model.encoder.layers.15.self_attn.out_proj", | |
| "vision_model.encoder.layers.15.mlp.fc1", | |
| "vision_model.encoder.layers.15.mlp.fc2", | |
| "vision_model.encoder.layers.16.self_attn.k_proj", | |
| "vision_model.encoder.layers.16.self_attn.v_proj", | |
| "vision_model.encoder.layers.16.self_attn.q_proj", | |
| "vision_model.encoder.layers.16.self_attn.out_proj", | |
| "vision_model.encoder.layers.16.mlp.fc1", | |
| "vision_model.encoder.layers.16.mlp.fc2", | |
| "vision_model.encoder.layers.17.self_attn.k_proj", | |
| "vision_model.encoder.layers.17.self_attn.v_proj", | |
| "vision_model.encoder.layers.17.self_attn.q_proj", | |
| "vision_model.encoder.layers.17.self_attn.out_proj", | |
| "vision_model.encoder.layers.17.mlp.fc1", | |
| "vision_model.encoder.layers.17.mlp.fc2", | |
| "vision_model.encoder.layers.18.self_attn.k_proj", | |
| "vision_model.encoder.layers.18.self_attn.v_proj", | |
| "vision_model.encoder.layers.18.self_attn.q_proj", | |
| "vision_model.encoder.layers.18.self_attn.out_proj", | |
| "vision_model.encoder.layers.18.mlp.fc1", | |
| "vision_model.encoder.layers.18.mlp.fc2", | |
| "vision_model.encoder.layers.19.self_attn.k_proj", | |
| "vision_model.encoder.layers.19.self_attn.v_proj", | |
| "vision_model.encoder.layers.19.self_attn.q_proj", | |
| "vision_model.encoder.layers.19.self_attn.out_proj", | |
| "vision_model.encoder.layers.19.mlp.fc1", | |
| "vision_model.encoder.layers.19.mlp.fc2", | |
| "vision_model.encoder.layers.20.self_attn.k_proj", | |
| "vision_model.encoder.layers.20.self_attn.v_proj", | |
| "vision_model.encoder.layers.20.self_attn.q_proj", | |
| "vision_model.encoder.layers.20.self_attn.out_proj", | |
| "vision_model.encoder.layers.20.mlp.fc1", | |
| "vision_model.encoder.layers.20.mlp.fc2", | |
| "vision_model.encoder.layers.21.self_attn.k_proj", | |
| "vision_model.encoder.layers.21.self_attn.v_proj", | |
| "vision_model.encoder.layers.21.self_attn.q_proj", | |
| "vision_model.encoder.layers.21.self_attn.out_proj", | |
| "vision_model.encoder.layers.21.mlp.fc1", | |
| "vision_model.encoder.layers.21.mlp.fc2", | |
| "vision_model.encoder.layers.22.self_attn.k_proj", | |
| "vision_model.encoder.layers.22.self_attn.v_proj", | |
| "vision_model.encoder.layers.22.self_attn.q_proj", | |
| "vision_model.encoder.layers.22.self_attn.out_proj", | |
| "vision_model.encoder.layers.22.mlp.fc1", | |
| "vision_model.encoder.layers.22.mlp.fc2", | |
| "vision_model.encoder.layers.23.self_attn.k_proj", | |
| "vision_model.encoder.layers.23.self_attn.v_proj", | |
| "vision_model.encoder.layers.23.self_attn.q_proj", | |
| "vision_model.encoder.layers.23.self_attn.out_proj", | |
| "vision_model.encoder.layers.23.mlp.fc1", | |
| "vision_model.encoder.layers.23.mlp.fc2", | |
| "vision_model.encoder.layers.24.self_attn.k_proj", | |
| "vision_model.encoder.layers.24.self_attn.v_proj", | |
| "vision_model.encoder.layers.24.self_attn.q_proj", | |
| "vision_model.encoder.layers.24.self_attn.out_proj", | |
| "vision_model.encoder.layers.24.mlp.fc1", | |
| "vision_model.encoder.layers.24.mlp.fc2", | |
| "vision_model.encoder.layers.25.self_attn.k_proj", | |
| "vision_model.encoder.layers.25.self_attn.v_proj", | |
| "vision_model.encoder.layers.25.self_attn.q_proj", | |
| "vision_model.encoder.layers.25.self_attn.out_proj", | |
| "vision_model.encoder.layers.25.mlp.fc1", | |
| "vision_model.encoder.layers.25.mlp.fc2", | |
| "vision_model.embeddings.patch_embedding" | |
| ] | |
| }, | |
| "threads_per_block": [ | |
| 512 | |
| ], | |
| "version": "0.2.0" | |
| }, | |
| "model_type": "qwen2" | |
| } | |