| { |
| "architectures": [ |
| "MiniMaxM3SparseForConditionalGeneration" |
| ], |
| "auto_map": { |
| "AutoConfig": "configuration_minimax_m3_vl.MiniMaxM3VLConfig" |
| }, |
| "eos_token_id": 200020, |
| "image_grid_pinpoints": "[(336, 336), (336, 672), (336, 1008), (336, 1344), (336, 1680), (336, 2016), (672, 336), (672, 672), (672, 1008), (672, 1344), (672, 1680), (672, 2016), (1008, 336), (1008, 672), (1008, 1008), (1008, 1344), (1008, 1680), (1008, 2016), (1344, 336), (1344, 672), (1344, 1008), (1344, 1344), (1344, 1680), (1344, 2016), (1680, 336), (1680, 672), (1680, 1008), (1680, 1344), (1680, 1680), (1680, 2016), (2016, 336), (2016, 672), (2016, 1008), (2016, 1344), (2016, 1680), (2016, 2016)]", |
| "image_seq_length": 576, |
| "image_token_index": 200025, |
| "img_token_compression_config": { |
| "image_token_compression_method": "patch_merge", |
| "spatial_merge_size": 2, |
| "temporal_patch_size": 2 |
| }, |
| "model_type": "minimax_m3_vl", |
| "multimodal_projector_bias": true, |
| "num_reward_heads": 0, |
| "process_image_mode": "dynamic_res", |
| "projector_hidden_act": "gelu", |
| "projector_hidden_size": 6144, |
| "quantization": { |
| "group_size": 64, |
| "bits": 4, |
| "mode": "affine", |
| "language_model.model.embed_tokens": { |
| "bits": 8, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.0.self_attn.q_proj": { |
| "bits": 8, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.0.self_attn.k_proj": { |
| "bits": 8, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.0.self_attn.v_proj": { |
| "bits": 8, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.0.self_attn.o_proj": { |
| "bits": 8, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.0.mlp.gate_proj": { |
| "bits": 8, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.0.mlp.up_proj": { |
| "bits": 8, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.0.mlp.down_proj": { |
| "bits": 8, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.1.self_attn.q_proj": { |
| "bits": 8, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.1.self_attn.k_proj": { |
| "bits": 8, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.1.self_attn.v_proj": { |
| "bits": 8, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.1.self_attn.o_proj": { |
| "bits": 8, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.1.mlp.gate_proj": { |
| "bits": 8, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.1.mlp.up_proj": { |
| "bits": 8, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.1.mlp.down_proj": { |
| "bits": 8, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.2.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.2.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.2.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.2.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.2.mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.2.mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.2.mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.3.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.3.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.3.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.3.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.3.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.3.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.3.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.3.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.3.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.3.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.4.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.4.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.4.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.4.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.4.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.4.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.4.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.4.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.4.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.4.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.5.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.5.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.5.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.5.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.5.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.5.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.5.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.5.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.5.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.5.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.6.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.6.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.6.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.6.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.6.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.6.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.6.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.6.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.6.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.6.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.7.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.7.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.7.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.7.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.7.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.7.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.7.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.7.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.7.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.7.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.8.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.8.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.8.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.8.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.8.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.8.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.8.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.8.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.8.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.8.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.9.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.9.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.9.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.9.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.9.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.9.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.9.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.9.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.9.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.9.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.10.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.10.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.10.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.10.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.10.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.10.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.10.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.10.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.10.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.10.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.11.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.11.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.11.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.11.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.11.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.11.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.11.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.11.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.11.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.11.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.12.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.12.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.12.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.12.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.12.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.12.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.12.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.12.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.12.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.12.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.13.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.13.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.13.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.13.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.13.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.13.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.13.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.13.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.13.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.13.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.14.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.14.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.14.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.14.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.14.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.14.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.14.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.14.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.14.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.14.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.15.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.15.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.15.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.15.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.15.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.15.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.15.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.15.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.15.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.15.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.16.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.16.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.16.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.16.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.16.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.16.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.16.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.16.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.16.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.16.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.17.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.17.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.17.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.17.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.17.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.17.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.17.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.17.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.17.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.17.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.18.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.18.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.18.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.18.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.18.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.18.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.18.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.18.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.18.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.18.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.19.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.19.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.19.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.19.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.19.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.19.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.19.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.19.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.19.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.19.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.20.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.20.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.20.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.20.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.20.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.20.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.20.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.20.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.20.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.20.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.21.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.21.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.21.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.21.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.21.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.21.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.21.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.21.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.21.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.21.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.22.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.22.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.22.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.22.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.22.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.22.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.22.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.22.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.22.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.22.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.23.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.23.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.23.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.23.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.23.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.23.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.23.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.23.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.23.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.23.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.24.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.24.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.24.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.24.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.24.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.24.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.24.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.24.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.24.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.24.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.25.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.25.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.25.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.25.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.25.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.25.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.25.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.25.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.25.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.25.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.26.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.26.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.26.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.26.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.26.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.26.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.26.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.26.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.26.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.26.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.27.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.27.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.27.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.27.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.27.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.27.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.27.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.27.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.27.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.27.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.28.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.28.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.28.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.28.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.28.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.28.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.28.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.28.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.28.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.28.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.29.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.29.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.29.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.29.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.29.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.29.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.29.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.29.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.29.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.29.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.30.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.30.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.30.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.30.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.30.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.30.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.30.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.30.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.30.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.30.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.31.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.31.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.31.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.31.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.31.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.31.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.31.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.31.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.31.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.31.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.32.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.32.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.32.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.32.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.32.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.32.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.32.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.32.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.32.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.32.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.33.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.33.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.33.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.33.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.33.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.33.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.33.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.33.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.33.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.33.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.34.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.34.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.34.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.34.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.34.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.34.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.34.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.34.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.34.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.34.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.35.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.35.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.35.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.35.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.35.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.35.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.35.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.35.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.35.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.35.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.36.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.36.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.36.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.36.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.36.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.36.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.36.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.36.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.36.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.36.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.37.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.37.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.37.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.37.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.37.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.37.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.37.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.37.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.37.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.37.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.38.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.38.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.38.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.38.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.38.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.38.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.38.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.38.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.38.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.38.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.39.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.39.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.39.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.39.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.39.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.39.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.39.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.39.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.39.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.39.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.40.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.40.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.40.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.40.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.40.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.40.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.40.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.40.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.40.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.40.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.41.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.41.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.41.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.41.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.41.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.41.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.41.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.41.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.41.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.41.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.42.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.42.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.42.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.42.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.42.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.42.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.42.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.42.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.42.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.42.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.43.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.43.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.43.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.43.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.43.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.43.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.43.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.43.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.43.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.43.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.44.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.44.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.44.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.44.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.44.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.44.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.44.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.44.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.44.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.44.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.45.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.45.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.45.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.45.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.45.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.45.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.45.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.45.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.45.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.45.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.46.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.46.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.46.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.46.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.46.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.46.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.46.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.46.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.46.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.46.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.47.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.47.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.47.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.47.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.47.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.47.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.47.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.47.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.47.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.47.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.48.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.48.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.48.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.48.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.48.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.48.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.48.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.48.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.48.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.48.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.49.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.49.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.49.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.49.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.49.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.49.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.49.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.49.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.49.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.49.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.50.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.50.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.50.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.50.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.50.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.50.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.50.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.50.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.50.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.50.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.51.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.51.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.51.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.51.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.51.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.51.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.51.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.51.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.51.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.51.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.52.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.52.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.52.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.52.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.52.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.52.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.52.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.52.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.52.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.52.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.53.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.53.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.53.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.53.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.53.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.53.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.53.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.53.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.53.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.53.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.54.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.54.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.54.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.54.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.54.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.54.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.54.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.54.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.54.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.54.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.55.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.55.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.55.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.55.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.55.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.55.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.55.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.55.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.55.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.55.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.56.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.56.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.56.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.56.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.56.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.56.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.56.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.56.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.56.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.56.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.57.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.57.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.57.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.57.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.57.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.57.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.57.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.57.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.57.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.57.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.58.self_attn.q_proj": { |
| "bits": 8, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.58.self_attn.k_proj": { |
| "bits": 8, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.58.self_attn.v_proj": { |
| "bits": 8, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.58.self_attn.o_proj": { |
| "bits": 8, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.58.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 8, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.58.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 8, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.58.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 8, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.58.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 8, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.58.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 8, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.58.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 8, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.59.self_attn.q_proj": { |
| "bits": 8, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.59.self_attn.k_proj": { |
| "bits": 8, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.59.self_attn.v_proj": { |
| "bits": 8, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.59.self_attn.o_proj": { |
| "bits": 8, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.59.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 8, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.59.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 8, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.59.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 8, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.59.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 8, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.59.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 8, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.59.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 8, |
| "group_size": 64 |
| }, |
| "language_model.lm_head": { |
| "bits": 8, |
| "group_size": 64 |
| } |
| }, |
| "quantization_config": { |
| "group_size": 64, |
| "bits": 4, |
| "mode": "affine", |
| "language_model.model.embed_tokens": { |
| "bits": 8, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.0.self_attn.q_proj": { |
| "bits": 8, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.0.self_attn.k_proj": { |
| "bits": 8, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.0.self_attn.v_proj": { |
| "bits": 8, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.0.self_attn.o_proj": { |
| "bits": 8, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.0.mlp.gate_proj": { |
| "bits": 8, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.0.mlp.up_proj": { |
| "bits": 8, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.0.mlp.down_proj": { |
| "bits": 8, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.1.self_attn.q_proj": { |
| "bits": 8, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.1.self_attn.k_proj": { |
| "bits": 8, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.1.self_attn.v_proj": { |
| "bits": 8, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.1.self_attn.o_proj": { |
| "bits": 8, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.1.mlp.gate_proj": { |
| "bits": 8, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.1.mlp.up_proj": { |
| "bits": 8, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.1.mlp.down_proj": { |
| "bits": 8, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.2.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.2.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.2.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.2.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.2.mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.2.mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.2.mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.3.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.3.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.3.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.3.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.3.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.3.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.3.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.3.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.3.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.3.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.4.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.4.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.4.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.4.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.4.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.4.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.4.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.4.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.4.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.4.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.5.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.5.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.5.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.5.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.5.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.5.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.5.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.5.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.5.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.5.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.6.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.6.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.6.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.6.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.6.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.6.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.6.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.6.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.6.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.6.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.7.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.7.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.7.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.7.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.7.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.7.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.7.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.7.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.7.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.7.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.8.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.8.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.8.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.8.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.8.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.8.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.8.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.8.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.8.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.8.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.9.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.9.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.9.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.9.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.9.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.9.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.9.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.9.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.9.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.9.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.10.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.10.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.10.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.10.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.10.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.10.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.10.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.10.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.10.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.10.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.11.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.11.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.11.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.11.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.11.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.11.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.11.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.11.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.11.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.11.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.12.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.12.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.12.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.12.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.12.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.12.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.12.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.12.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.12.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.12.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.13.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.13.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.13.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.13.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.13.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.13.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.13.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.13.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.13.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.13.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.14.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.14.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.14.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.14.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.14.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.14.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.14.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.14.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.14.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.14.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.15.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.15.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.15.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.15.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.15.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.15.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.15.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.15.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.15.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.15.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.16.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.16.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.16.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.16.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.16.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.16.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.16.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.16.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.16.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.16.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.17.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.17.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.17.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.17.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.17.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.17.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.17.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.17.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.17.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.17.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.18.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.18.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.18.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.18.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.18.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.18.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.18.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.18.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.18.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.18.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.19.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.19.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.19.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.19.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.19.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.19.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.19.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.19.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.19.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.19.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.20.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.20.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.20.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.20.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.20.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.20.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.20.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.20.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.20.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.20.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.21.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.21.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.21.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.21.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.21.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.21.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.21.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.21.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.21.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.21.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.22.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.22.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.22.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.22.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.22.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.22.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.22.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.22.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.22.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.22.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.23.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.23.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.23.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.23.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.23.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.23.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.23.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.23.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.23.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.23.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.24.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.24.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.24.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.24.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.24.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.24.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.24.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.24.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.24.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.24.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.25.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.25.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.25.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.25.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.25.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.25.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.25.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.25.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.25.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.25.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.26.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.26.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.26.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.26.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.26.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.26.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.26.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.26.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.26.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.26.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.27.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.27.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.27.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.27.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.27.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.27.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.27.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.27.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.27.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.27.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.28.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.28.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.28.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.28.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.28.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.28.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.28.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.28.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.28.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.28.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.29.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.29.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.29.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.29.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.29.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.29.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.29.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.29.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.29.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.29.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.30.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.30.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.30.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.30.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.30.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.30.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.30.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.30.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.30.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.30.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.31.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.31.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.31.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.31.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.31.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.31.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.31.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.31.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.31.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.31.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.32.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.32.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.32.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.32.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.32.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.32.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.32.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.32.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.32.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.32.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.33.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.33.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.33.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.33.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.33.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.33.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.33.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.33.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.33.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.33.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.34.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.34.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.34.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.34.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.34.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.34.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.34.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.34.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.34.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.34.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.35.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.35.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.35.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.35.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.35.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.35.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.35.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.35.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.35.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.35.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.36.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.36.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.36.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.36.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.36.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.36.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.36.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.36.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.36.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.36.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.37.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.37.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.37.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.37.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.37.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.37.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.37.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.37.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.37.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.37.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.38.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.38.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.38.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.38.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.38.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.38.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.38.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.38.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.38.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.38.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.39.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.39.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.39.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.39.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.39.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.39.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.39.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.39.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.39.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.39.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.40.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.40.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.40.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.40.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.40.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.40.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.40.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.40.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.40.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.40.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.41.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.41.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.41.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.41.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.41.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.41.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.41.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.41.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.41.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.41.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.42.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.42.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.42.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.42.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.42.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.42.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.42.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.42.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.42.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.42.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.43.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.43.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.43.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.43.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.43.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.43.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.43.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.43.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.43.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.43.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.44.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.44.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.44.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.44.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.44.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.44.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.44.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.44.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.44.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.44.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.45.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.45.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.45.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.45.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.45.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.45.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.45.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.45.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.45.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.45.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.46.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.46.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.46.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.46.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.46.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.46.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.46.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.46.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.46.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.46.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.47.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.47.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.47.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.47.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.47.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.47.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.47.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.47.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.47.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.47.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.48.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.48.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.48.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.48.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.48.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.48.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.48.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.48.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.48.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.48.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.49.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.49.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.49.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.49.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.49.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.49.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.49.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.49.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.49.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.49.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.50.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.50.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.50.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.50.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.50.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.50.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.50.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.50.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.50.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.50.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.51.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.51.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.51.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.51.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.51.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.51.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.51.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.51.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.51.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.51.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.52.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.52.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.52.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.52.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.52.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.52.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.52.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.52.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.52.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.52.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.53.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.53.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.53.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.53.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.53.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.53.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.53.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.53.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.53.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.53.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.54.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.54.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.54.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.54.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.54.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.54.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.54.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.54.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.54.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.54.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.55.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.55.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.55.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.55.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.55.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.55.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.55.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.55.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.55.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.55.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.56.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.56.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.56.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.56.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.56.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.56.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.56.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.56.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.56.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.56.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.57.self_attn.q_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.57.self_attn.k_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.57.self_attn.v_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.57.self_attn.o_proj": { |
| "bits": 4, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.57.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.57.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.57.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.57.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.57.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 2, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.57.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 3, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.58.self_attn.q_proj": { |
| "bits": 8, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.58.self_attn.k_proj": { |
| "bits": 8, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.58.self_attn.v_proj": { |
| "bits": 8, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.58.self_attn.o_proj": { |
| "bits": 8, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.58.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 8, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.58.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 8, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.58.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 8, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.58.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 8, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.58.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 8, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.58.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 8, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.59.self_attn.q_proj": { |
| "bits": 8, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.59.self_attn.k_proj": { |
| "bits": 8, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.59.self_attn.v_proj": { |
| "bits": 8, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.59.self_attn.o_proj": { |
| "bits": 8, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.59.block_sparse_moe.switch_mlp.gate_proj": { |
| "bits": 8, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.59.block_sparse_moe.switch_mlp.up_proj": { |
| "bits": 8, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.59.block_sparse_moe.switch_mlp.down_proj": { |
| "bits": 8, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.59.block_sparse_moe.shared_experts.gate_proj": { |
| "bits": 8, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.59.block_sparse_moe.shared_experts.up_proj": { |
| "bits": 8, |
| "group_size": 64 |
| }, |
| "language_model.model.layers.59.block_sparse_moe.shared_experts.down_proj": { |
| "bits": 8, |
| "group_size": 64 |
| }, |
| "language_model.lm_head": { |
| "bits": 8, |
| "group_size": 64 |
| } |
| }, |
| "text_config": { |
| "dtype": "bfloat16", |
| "hidden_size": 6144, |
| "intermediate_size": 3072, |
| "num_hidden_layers": 60, |
| "num_attention_heads": 64, |
| "num_key_value_heads": 4, |
| "head_dim": 128, |
| "vocab_size": 200064, |
| "max_position_embeddings": 1048576, |
| "rms_norm_eps": 1e-06, |
| "use_gemma_norm": true, |
| "attention_output_gate": false, |
| "rope_theta": 5000000, |
| "rotary_dim": 64, |
| "partial_rotary_factor": 0.5, |
| "hidden_act": "swigluoai", |
| "use_qk_norm": true, |
| "tie_word_embeddings": false, |
| "dense_intermediate_size": 12288, |
| "shared_intermediate_size": 3072, |
| "num_local_experts": 128, |
| "num_experts_per_tok": 4, |
| "n_shared_experts": 1, |
| "scoring_func": "sigmoid", |
| "use_routing_bias": true, |
| "moe_layer_freq": [ |
| 0, |
| 0, |
| 0, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1 |
| ], |
| "qk_norm_type": "per_head", |
| "num_mtp_modules": 1, |
| "swiglu_alpha": 1.702, |
| "swiglu_limit": 7.0, |
| "routed_scaling_factor": 2.0, |
| "sparse_attention_config": { |
| "use_sparse_attention": true, |
| "sparse_index_dim": 128, |
| "sparse_num_index_heads": 4, |
| "sparse_topk_blocks": 16, |
| "sparse_block_size": 128, |
| "sparse_disable_index_value": [ |
| 0, |
| 0, |
| 0, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1 |
| ], |
| "sparse_score_type": "max", |
| "sparse_init_block": 0, |
| "sparse_local_block": 1, |
| "sparse_attention_freq": [ |
| 0, |
| 0, |
| 0, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1, |
| 1 |
| ] |
| }, |
| "architectures": [ |
| "MiniMaxM3SparseForCausalLM" |
| ] |
| }, |
| "torch_dtype": "bfloat16", |
| "transformers_version": "4.52.4", |
| "video_token_index": 200026, |
| "vision_feature_layer": -1, |
| "vision_feature_select_strategy": "full" |
| } |
