{ "thinker.model.language_model.embed_tokens.weight": [ 248320, 2048 ], "thinker.model.language_model.layers.0.linear_attn.dt_bias": [ 16 ], "thinker.model.language_model.layers.0.linear_attn.A_log": [ 16 ], "thinker.model.language_model.layers.0.linear_attn.conv1d.weight": [ 6144, 1, 4 ], "thinker.model.language_model.layers.0.linear_attn.norm.weight": [ 128 ], "thinker.model.language_model.layers.0.linear_attn.out_proj.weight": [ 2048, 2048 ], "thinker.model.language_model.layers.0.linear_attn.in_proj_qkv.weight": [ 6144, 2048 ], "thinker.model.language_model.layers.0.linear_attn.in_proj_z.weight": [ 2048, 2048 ], "thinker.model.language_model.layers.0.linear_attn.in_proj_b.weight": [ 16, 2048 ], "thinker.model.language_model.layers.0.linear_attn.in_proj_a.weight": [ 16, 2048 ], "thinker.model.language_model.layers.0.mlp.gate_proj.weight": [ 6144, 2048 ], "thinker.model.language_model.layers.0.mlp.up_proj.weight": [ 6144, 2048 ], "thinker.model.language_model.layers.0.mlp.down_proj.weight": [ 2048, 6144 ], "thinker.model.language_model.layers.0.input_layernorm.weight": [ 2048 ], "thinker.model.language_model.layers.0.post_attention_layernorm.weight": [ 2048 ], "thinker.model.language_model.layers.1.linear_attn.dt_bias": [ 16 ], "thinker.model.language_model.layers.1.linear_attn.A_log": [ 16 ], "thinker.model.language_model.layers.1.linear_attn.conv1d.weight": [ 6144, 1, 4 ], "thinker.model.language_model.layers.1.linear_attn.norm.weight": [ 128 ], "thinker.model.language_model.layers.1.linear_attn.out_proj.weight": [ 2048, 2048 ], "thinker.model.language_model.layers.1.linear_attn.in_proj_qkv.weight": [ 6144, 2048 ], "thinker.model.language_model.layers.1.linear_attn.in_proj_z.weight": [ 2048, 2048 ], "thinker.model.language_model.layers.1.linear_attn.in_proj_b.weight": [ 16, 2048 ], "thinker.model.language_model.layers.1.linear_attn.in_proj_a.weight": [ 16, 2048 ], "thinker.model.language_model.layers.1.mlp.gate_proj.weight": [ 6144, 2048 ], "thinker.model.language_model.layers.1.mlp.up_proj.weight": [ 6144, 2048 ], "thinker.model.language_model.layers.1.mlp.down_proj.weight": [ 2048, 6144 ], "thinker.model.language_model.layers.1.input_layernorm.weight": [ 2048 ], "thinker.model.language_model.layers.1.post_attention_layernorm.weight": [ 2048 ], "thinker.model.language_model.layers.2.linear_attn.dt_bias": [ 16 ], "thinker.model.language_model.layers.2.linear_attn.A_log": [ 16 ], "thinker.model.language_model.layers.2.linear_attn.conv1d.weight": [ 6144, 1, 4 ], "thinker.model.language_model.layers.2.linear_attn.norm.weight": [ 128 ], "thinker.model.language_model.layers.2.linear_attn.out_proj.weight": [ 2048, 2048 ], "thinker.model.language_model.layers.2.linear_attn.in_proj_qkv.weight": [ 6144, 2048 ], "thinker.model.language_model.layers.2.linear_attn.in_proj_z.weight": [ 2048, 2048 ], "thinker.model.language_model.layers.2.linear_attn.in_proj_b.weight": [ 16, 2048 ], "thinker.model.language_model.layers.2.linear_attn.in_proj_a.weight": [ 16, 2048 ], "thinker.model.language_model.layers.2.mlp.gate_proj.weight": [ 6144, 2048 ], "thinker.model.language_model.layers.2.mlp.up_proj.weight": [ 6144, 2048 ], "thinker.model.language_model.layers.2.mlp.down_proj.weight": [ 2048, 6144 ], "thinker.model.language_model.layers.2.input_layernorm.weight": [ 2048 ], "thinker.model.language_model.layers.2.post_attention_layernorm.weight": [ 2048 ], "thinker.model.language_model.layers.3.self_attn.q_proj.weight": [ 4096, 2048 ], "thinker.model.language_model.layers.3.self_attn.k_proj.weight": [ 512, 2048 ], "thinker.model.language_model.layers.3.self_attn.v_proj.weight": [ 512, 2048 ], "thinker.model.language_model.layers.3.self_attn.o_proj.weight": [ 2048, 2048 ], "thinker.model.language_model.layers.3.self_attn.q_norm.weight": [ 256 ], "thinker.model.language_model.layers.3.self_attn.k_norm.weight": [ 256 ], "thinker.model.language_model.layers.3.mlp.gate_proj.weight": [ 6144, 2048 ], "thinker.model.language_model.layers.3.mlp.up_proj.weight": [ 6144, 2048 ], "thinker.model.language_model.layers.3.mlp.down_proj.weight": [ 2048, 6144 ], "thinker.model.language_model.layers.3.input_layernorm.weight": [ 2048 ], "thinker.model.language_model.layers.3.post_attention_layernorm.weight": [ 2048 ], "thinker.model.language_model.layers.4.linear_attn.dt_bias": [ 16 ], "thinker.model.language_model.layers.4.linear_attn.A_log": [ 16 ], "thinker.model.language_model.layers.4.linear_attn.conv1d.weight": [ 6144, 1, 4 ], "thinker.model.language_model.layers.4.linear_attn.norm.weight": [ 128 ], "thinker.model.language_model.layers.4.linear_attn.out_proj.weight": [ 2048, 2048 ], "thinker.model.language_model.layers.4.linear_attn.in_proj_qkv.weight": [ 6144, 2048 ], "thinker.model.language_model.layers.4.linear_attn.in_proj_z.weight": [ 2048, 2048 ], "thinker.model.language_model.layers.4.linear_attn.in_proj_b.weight": [ 16, 2048 ], "thinker.model.language_model.layers.4.linear_attn.in_proj_a.weight": [ 16, 2048 ], "thinker.model.language_model.layers.4.mlp.gate_proj.weight": [ 6144, 2048 ], "thinker.model.language_model.layers.4.mlp.up_proj.weight": [ 6144, 2048 ], "thinker.model.language_model.layers.4.mlp.down_proj.weight": [ 2048, 6144 ], "thinker.model.language_model.layers.4.input_layernorm.weight": [ 2048 ], "thinker.model.language_model.layers.4.post_attention_layernorm.weight": [ 2048 ], "thinker.model.language_model.layers.5.linear_attn.dt_bias": [ 16 ], "thinker.model.language_model.layers.5.linear_attn.A_log": [ 16 ], "thinker.model.language_model.layers.5.linear_attn.conv1d.weight": [ 6144, 1, 4 ], "thinker.model.language_model.layers.5.linear_attn.norm.weight": [ 128 ], "thinker.model.language_model.layers.5.linear_attn.out_proj.weight": [ 2048, 2048 ], "thinker.model.language_model.layers.5.linear_attn.in_proj_qkv.weight": [ 6144, 2048 ], "thinker.model.language_model.layers.5.linear_attn.in_proj_z.weight": [ 2048, 2048 ], "thinker.model.language_model.layers.5.linear_attn.in_proj_b.weight": [ 16, 2048 ], "thinker.model.language_model.layers.5.linear_attn.in_proj_a.weight": [ 16, 2048 ], "thinker.model.language_model.layers.5.mlp.gate_proj.weight": [ 6144, 2048 ], "thinker.model.language_model.layers.5.mlp.up_proj.weight": [ 6144, 2048 ], "thinker.model.language_model.layers.5.mlp.down_proj.weight": [ 2048, 6144 ], "thinker.model.language_model.layers.5.input_layernorm.weight": [ 2048 ], "thinker.model.language_model.layers.5.post_attention_layernorm.weight": [ 2048 ], "thinker.model.language_model.layers.6.linear_attn.dt_bias": [ 16 ], "thinker.model.language_model.layers.6.linear_attn.A_log": [ 16 ], "thinker.model.language_model.layers.6.linear_attn.conv1d.weight": [ 6144, 1, 4 ], "thinker.model.language_model.layers.6.linear_attn.norm.weight": [ 128 ], "thinker.model.language_model.layers.6.linear_attn.out_proj.weight": [ 2048, 2048 ], "thinker.model.language_model.layers.6.linear_attn.in_proj_qkv.weight": [ 6144, 2048 ], "thinker.model.language_model.layers.6.linear_attn.in_proj_z.weight": [ 2048, 2048 ], "thinker.model.language_model.layers.6.linear_attn.in_proj_b.weight": [ 16, 2048 ], "thinker.model.language_model.layers.6.linear_attn.in_proj_a.weight": [ 16, 2048 ], "thinker.model.language_model.layers.6.mlp.gate_proj.weight": [ 6144, 2048 ], "thinker.model.language_model.layers.6.mlp.up_proj.weight": [ 6144, 2048 ], "thinker.model.language_model.layers.6.mlp.down_proj.weight": [ 2048, 6144 ], "thinker.model.language_model.layers.6.input_layernorm.weight": [ 2048 ], "thinker.model.language_model.layers.6.post_attention_layernorm.weight": [ 2048 ], "thinker.model.language_model.layers.7.self_attn.q_proj.weight": [ 4096, 2048 ], "thinker.model.language_model.layers.7.self_attn.k_proj.weight": [ 512, 2048 ], "thinker.model.language_model.layers.7.self_attn.v_proj.weight": [ 512, 2048 ], "thinker.model.language_model.layers.7.self_attn.o_proj.weight": [ 2048, 2048 ], "thinker.model.language_model.layers.7.self_attn.q_norm.weight": [ 256 ], "thinker.model.language_model.layers.7.self_attn.k_norm.weight": [ 256 ], "thinker.model.language_model.layers.7.mlp.gate_proj.weight": [ 6144, 2048 ], "thinker.model.language_model.layers.7.mlp.up_proj.weight": [ 6144, 2048 ], "thinker.model.language_model.layers.7.mlp.down_proj.weight": [ 2048, 6144 ], "thinker.model.language_model.layers.7.input_layernorm.weight": [ 2048 ], "thinker.model.language_model.layers.7.post_attention_layernorm.weight": [ 2048 ], "thinker.model.language_model.layers.8.linear_attn.dt_bias": [ 16 ], "thinker.model.language_model.layers.8.linear_attn.A_log": [ 16 ], "thinker.model.language_model.layers.8.linear_attn.conv1d.weight": [ 6144, 1, 4 ], "thinker.model.language_model.layers.8.linear_attn.norm.weight": [ 128 ], "thinker.model.language_model.layers.8.linear_attn.out_proj.weight": [ 2048, 2048 ], "thinker.model.language_model.layers.8.linear_attn.in_proj_qkv.weight": [ 6144, 2048 ], "thinker.model.language_model.layers.8.linear_attn.in_proj_z.weight": [ 2048, 2048 ], "thinker.model.language_model.layers.8.linear_attn.in_proj_b.weight": [ 16, 2048 ], "thinker.model.language_model.layers.8.linear_attn.in_proj_a.weight": [ 16, 2048 ], "thinker.model.language_model.layers.8.mlp.gate_proj.weight": [ 6144, 2048 ], "thinker.model.language_model.layers.8.mlp.up_proj.weight": [ 6144, 2048 ], "thinker.model.language_model.layers.8.mlp.down_proj.weight": [ 2048, 6144 ], "thinker.model.language_model.layers.8.input_layernorm.weight": [ 2048 ], "thinker.model.language_model.layers.8.post_attention_layernorm.weight": [ 2048 ], "thinker.model.language_model.layers.9.linear_attn.dt_bias": [ 16 ], "thinker.model.language_model.layers.9.linear_attn.A_log": [ 16 ], "thinker.model.language_model.layers.9.linear_attn.conv1d.weight": [ 6144, 1, 4 ], "thinker.model.language_model.layers.9.linear_attn.norm.weight": [ 128 ], "thinker.model.language_model.layers.9.linear_attn.out_proj.weight": [ 2048, 2048 ], "thinker.model.language_model.layers.9.linear_attn.in_proj_qkv.weight": [ 6144, 2048 ], "thinker.model.language_model.layers.9.linear_attn.in_proj_z.weight": [ 2048, 2048 ], "thinker.model.language_model.layers.9.linear_attn.in_proj_b.weight": [ 16, 2048 ], "thinker.model.language_model.layers.9.linear_attn.in_proj_a.weight": [ 16, 2048 ], "thinker.model.language_model.layers.9.mlp.gate_proj.weight": [ 6144, 2048 ], "thinker.model.language_model.layers.9.mlp.up_proj.weight": [ 6144, 2048 ], "thinker.model.language_model.layers.9.mlp.down_proj.weight": [ 2048, 6144 ], "thinker.model.language_model.layers.9.input_layernorm.weight": [ 2048 ], "thinker.model.language_model.layers.9.post_attention_layernorm.weight": [ 2048 ], "thinker.model.language_model.layers.10.linear_attn.dt_bias": [ 16 ], "thinker.model.language_model.layers.10.linear_attn.A_log": [ 16 ], "thinker.model.language_model.layers.10.linear_attn.conv1d.weight": [ 6144, 1, 4 ], "thinker.model.language_model.layers.10.linear_attn.norm.weight": [ 128 ], "thinker.model.language_model.layers.10.linear_attn.out_proj.weight": [ 2048, 2048 ], "thinker.model.language_model.layers.10.linear_attn.in_proj_qkv.weight": [ 6144, 2048 ], "thinker.model.language_model.layers.10.linear_attn.in_proj_z.weight": [ 2048, 2048 ], "thinker.model.language_model.layers.10.linear_attn.in_proj_b.weight": [ 16, 2048 ], "thinker.model.language_model.layers.10.linear_attn.in_proj_a.weight": [ 16, 2048 ], "thinker.model.language_model.layers.10.mlp.gate_proj.weight": [ 6144, 2048 ], "thinker.model.language_model.layers.10.mlp.up_proj.weight": [ 6144, 2048 ], "thinker.model.language_model.layers.10.mlp.down_proj.weight": [ 2048, 6144 ], "thinker.model.language_model.layers.10.input_layernorm.weight": [ 2048 ], "thinker.model.language_model.layers.10.post_attention_layernorm.weight": [ 2048 ], "thinker.model.language_model.layers.11.self_attn.q_proj.weight": [ 4096, 2048 ], "thinker.model.language_model.layers.11.self_attn.k_proj.weight": [ 512, 2048 ], "thinker.model.language_model.layers.11.self_attn.v_proj.weight": [ 512, 2048 ], "thinker.model.language_model.layers.11.self_attn.o_proj.weight": [ 2048, 2048 ], "thinker.model.language_model.layers.11.self_attn.q_norm.weight": [ 256 ], "thinker.model.language_model.layers.11.self_attn.k_norm.weight": [ 256 ], "thinker.model.language_model.layers.11.mlp.gate_proj.weight": [ 6144, 2048 ], "thinker.model.language_model.layers.11.mlp.up_proj.weight": [ 6144, 2048 ], "thinker.model.language_model.layers.11.mlp.down_proj.weight": [ 2048, 6144 ], "thinker.model.language_model.layers.11.input_layernorm.weight": [ 2048 ], "thinker.model.language_model.layers.11.post_attention_layernorm.weight": [ 2048 ], "thinker.model.language_model.layers.12.linear_attn.dt_bias": [ 16 ], "thinker.model.language_model.layers.12.linear_attn.A_log": [ 16 ], "thinker.model.language_model.layers.12.linear_attn.conv1d.weight": [ 6144, 1, 4 ], "thinker.model.language_model.layers.12.linear_attn.norm.weight": [ 128 ], "thinker.model.language_model.layers.12.linear_attn.out_proj.weight": [ 2048, 2048 ], "thinker.model.language_model.layers.12.linear_attn.in_proj_qkv.weight": [ 6144, 2048 ], "thinker.model.language_model.layers.12.linear_attn.in_proj_z.weight": [ 2048, 2048 ], "thinker.model.language_model.layers.12.linear_attn.in_proj_b.weight": [ 16, 2048 ], "thinker.model.language_model.layers.12.linear_attn.in_proj_a.weight": [ 16, 2048 ], "thinker.model.language_model.layers.12.mlp.gate_proj.weight": [ 6144, 2048 ], "thinker.model.language_model.layers.12.mlp.up_proj.weight": [ 6144, 2048 ], "thinker.model.language_model.layers.12.mlp.down_proj.weight": [ 2048, 6144 ], "thinker.model.language_model.layers.12.input_layernorm.weight": [ 2048 ], "thinker.model.language_model.layers.12.post_attention_layernorm.weight": [ 2048 ], "thinker.model.language_model.layers.13.linear_attn.dt_bias": [ 16 ], "thinker.model.language_model.layers.13.linear_attn.A_log": [ 16 ], "thinker.model.language_model.layers.13.linear_attn.conv1d.weight": [ 6144, 1, 4 ], "thinker.model.language_model.layers.13.linear_attn.norm.weight": [ 128 ], "thinker.model.language_model.layers.13.linear_attn.out_proj.weight": [ 2048, 2048 ], "thinker.model.language_model.layers.13.linear_attn.in_proj_qkv.weight": [ 6144, 2048 ], "thinker.model.language_model.layers.13.linear_attn.in_proj_z.weight": [ 2048, 2048 ], "thinker.model.language_model.layers.13.linear_attn.in_proj_b.weight": [ 16, 2048 ], "thinker.model.language_model.layers.13.linear_attn.in_proj_a.weight": [ 16, 2048 ], "thinker.model.language_model.layers.13.mlp.gate_proj.weight": [ 6144, 2048 ], "thinker.model.language_model.layers.13.mlp.up_proj.weight": [ 6144, 2048 ], "thinker.model.language_model.layers.13.mlp.down_proj.weight": [ 2048, 6144 ], "thinker.model.language_model.layers.13.input_layernorm.weight": [ 2048 ], "thinker.model.language_model.layers.13.post_attention_layernorm.weight": [ 2048 ], "thinker.model.language_model.layers.14.linear_attn.dt_bias": [ 16 ], "thinker.model.language_model.layers.14.linear_attn.A_log": [ 16 ], "thinker.model.language_model.layers.14.linear_attn.conv1d.weight": [ 6144, 1, 4 ], "thinker.model.language_model.layers.14.linear_attn.norm.weight": [ 128 ], "thinker.model.language_model.layers.14.linear_attn.out_proj.weight": [ 2048, 2048 ], "thinker.model.language_model.layers.14.linear_attn.in_proj_qkv.weight": [ 6144, 2048 ], "thinker.model.language_model.layers.14.linear_attn.in_proj_z.weight": [ 2048, 2048 ], "thinker.model.language_model.layers.14.linear_attn.in_proj_b.weight": [ 16, 2048 ], "thinker.model.language_model.layers.14.linear_attn.in_proj_a.weight": [ 16, 2048 ], "thinker.model.language_model.layers.14.mlp.gate_proj.weight": [ 6144, 2048 ], "thinker.model.language_model.layers.14.mlp.up_proj.weight": [ 6144, 2048 ], "thinker.model.language_model.layers.14.mlp.down_proj.weight": [ 2048, 6144 ], "thinker.model.language_model.layers.14.input_layernorm.weight": [ 2048 ], "thinker.model.language_model.layers.14.post_attention_layernorm.weight": [ 2048 ], "thinker.model.language_model.layers.15.self_attn.q_proj.weight": [ 4096, 2048 ], "thinker.model.language_model.layers.15.self_attn.k_proj.weight": [ 512, 2048 ], "thinker.model.language_model.layers.15.self_attn.v_proj.weight": [ 512, 2048 ], "thinker.model.language_model.layers.15.self_attn.o_proj.weight": [ 2048, 2048 ], "thinker.model.language_model.layers.15.self_attn.q_norm.weight": [ 256 ], "thinker.model.language_model.layers.15.self_attn.k_norm.weight": [ 256 ], "thinker.model.language_model.layers.15.mlp.gate_proj.weight": [ 6144, 2048 ], "thinker.model.language_model.layers.15.mlp.up_proj.weight": [ 6144, 2048 ], "thinker.model.language_model.layers.15.mlp.down_proj.weight": [ 2048, 6144 ], "thinker.model.language_model.layers.15.input_layernorm.weight": [ 2048 ], "thinker.model.language_model.layers.15.post_attention_layernorm.weight": [ 2048 ], "thinker.model.language_model.layers.16.linear_attn.dt_bias": [ 16 ], "thinker.model.language_model.layers.16.linear_attn.A_log": [ 16 ], "thinker.model.language_model.layers.16.linear_attn.conv1d.weight": [ 6144, 1, 4 ], "thinker.model.language_model.layers.16.linear_attn.norm.weight": [ 128 ], "thinker.model.language_model.layers.16.linear_attn.out_proj.weight": [ 2048, 2048 ], "thinker.model.language_model.layers.16.linear_attn.in_proj_qkv.weight": [ 6144, 2048 ], "thinker.model.language_model.layers.16.linear_attn.in_proj_z.weight": [ 2048, 2048 ], "thinker.model.language_model.layers.16.linear_attn.in_proj_b.weight": [ 16, 2048 ], "thinker.model.language_model.layers.16.linear_attn.in_proj_a.weight": [ 16, 2048 ], "thinker.model.language_model.layers.16.mlp.gate_proj.weight": [ 6144, 2048 ], "thinker.model.language_model.layers.16.mlp.up_proj.weight": [ 6144, 2048 ], "thinker.model.language_model.layers.16.mlp.down_proj.weight": [ 2048, 6144 ], "thinker.model.language_model.layers.16.input_layernorm.weight": [ 2048 ], "thinker.model.language_model.layers.16.post_attention_layernorm.weight": [ 2048 ], "thinker.model.language_model.layers.17.linear_attn.dt_bias": [ 16 ], "thinker.model.language_model.layers.17.linear_attn.A_log": [ 16 ], "thinker.model.language_model.layers.17.linear_attn.conv1d.weight": [ 6144, 1, 4 ], "thinker.model.language_model.layers.17.linear_attn.norm.weight": [ 128 ], "thinker.model.language_model.layers.17.linear_attn.out_proj.weight": [ 2048, 2048 ], "thinker.model.language_model.layers.17.linear_attn.in_proj_qkv.weight": [ 6144, 2048 ], "thinker.model.language_model.layers.17.linear_attn.in_proj_z.weight": [ 2048, 2048 ], "thinker.model.language_model.layers.17.linear_attn.in_proj_b.weight": [ 16, 2048 ], "thinker.model.language_model.layers.17.linear_attn.in_proj_a.weight": [ 16, 2048 ], "thinker.model.language_model.layers.17.mlp.gate_proj.weight": [ 6144, 2048 ], "thinker.model.language_model.layers.17.mlp.up_proj.weight": [ 6144, 2048 ], "thinker.model.language_model.layers.17.mlp.down_proj.weight": [ 2048, 6144 ], "thinker.model.language_model.layers.17.input_layernorm.weight": [ 2048 ], "thinker.model.language_model.layers.17.post_attention_layernorm.weight": [ 2048 ], "thinker.model.language_model.layers.18.linear_attn.dt_bias": [ 16 ], "thinker.model.language_model.layers.18.linear_attn.A_log": [ 16 ], "thinker.model.language_model.layers.18.linear_attn.conv1d.weight": [ 6144, 1, 4 ], "thinker.model.language_model.layers.18.linear_attn.norm.weight": [ 128 ], "thinker.model.language_model.layers.18.linear_attn.out_proj.weight": [ 2048, 2048 ], "thinker.model.language_model.layers.18.linear_attn.in_proj_qkv.weight": [ 6144, 2048 ], "thinker.model.language_model.layers.18.linear_attn.in_proj_z.weight": [ 2048, 2048 ], "thinker.model.language_model.layers.18.linear_attn.in_proj_b.weight": [ 16, 2048 ], "thinker.model.language_model.layers.18.linear_attn.in_proj_a.weight": [ 16, 2048 ], "thinker.model.language_model.layers.18.mlp.gate_proj.weight": [ 6144, 2048 ], "thinker.model.language_model.layers.18.mlp.up_proj.weight": [ 6144, 2048 ], "thinker.model.language_model.layers.18.mlp.down_proj.weight": [ 2048, 6144 ], "thinker.model.language_model.layers.18.input_layernorm.weight": [ 2048 ], "thinker.model.language_model.layers.18.post_attention_layernorm.weight": [ 2048 ], "thinker.model.language_model.layers.19.self_attn.q_proj.weight": [ 4096, 2048 ], "thinker.model.language_model.layers.19.self_attn.k_proj.weight": [ 512, 2048 ], "thinker.model.language_model.layers.19.self_attn.v_proj.weight": [ 512, 2048 ], "thinker.model.language_model.layers.19.self_attn.o_proj.weight": [ 2048, 2048 ], "thinker.model.language_model.layers.19.self_attn.q_norm.weight": [ 256 ], "thinker.model.language_model.layers.19.self_attn.k_norm.weight": [ 256 ], "thinker.model.language_model.layers.19.mlp.gate_proj.weight": [ 6144, 2048 ], "thinker.model.language_model.layers.19.mlp.up_proj.weight": [ 6144, 2048 ], "thinker.model.language_model.layers.19.mlp.down_proj.weight": [ 2048, 6144 ], "thinker.model.language_model.layers.19.input_layernorm.weight": [ 2048 ], "thinker.model.language_model.layers.19.post_attention_layernorm.weight": [ 2048 ], "thinker.model.language_model.layers.20.linear_attn.dt_bias": [ 16 ], "thinker.model.language_model.layers.20.linear_attn.A_log": [ 16 ], "thinker.model.language_model.layers.20.linear_attn.conv1d.weight": [ 6144, 1, 4 ], "thinker.model.language_model.layers.20.linear_attn.norm.weight": [ 128 ], "thinker.model.language_model.layers.20.linear_attn.out_proj.weight": [ 2048, 2048 ], "thinker.model.language_model.layers.20.linear_attn.in_proj_qkv.weight": [ 6144, 2048 ], "thinker.model.language_model.layers.20.linear_attn.in_proj_z.weight": [ 2048, 2048 ], "thinker.model.language_model.layers.20.linear_attn.in_proj_b.weight": [ 16, 2048 ], "thinker.model.language_model.layers.20.linear_attn.in_proj_a.weight": [ 16, 2048 ], "thinker.model.language_model.layers.20.mlp.gate_proj.weight": [ 6144, 2048 ], "thinker.model.language_model.layers.20.mlp.up_proj.weight": [ 6144, 2048 ], "thinker.model.language_model.layers.20.mlp.down_proj.weight": [ 2048, 6144 ], "thinker.model.language_model.layers.20.input_layernorm.weight": [ 2048 ], "thinker.model.language_model.layers.20.post_attention_layernorm.weight": [ 2048 ], "thinker.model.language_model.layers.21.linear_attn.dt_bias": [ 16 ], "thinker.model.language_model.layers.21.linear_attn.A_log": [ 16 ], "thinker.model.language_model.layers.21.linear_attn.conv1d.weight": [ 6144, 1, 4 ], "thinker.model.language_model.layers.21.linear_attn.norm.weight": [ 128 ], "thinker.model.language_model.layers.21.linear_attn.out_proj.weight": [ 2048, 2048 ], "thinker.model.language_model.layers.21.linear_attn.in_proj_qkv.weight": [ 6144, 2048 ], "thinker.model.language_model.layers.21.linear_attn.in_proj_z.weight": [ 2048, 2048 ], "thinker.model.language_model.layers.21.linear_attn.in_proj_b.weight": [ 16, 2048 ], "thinker.model.language_model.layers.21.linear_attn.in_proj_a.weight": [ 16, 2048 ], "thinker.model.language_model.layers.21.mlp.gate_proj.weight": [ 6144, 2048 ], "thinker.model.language_model.layers.21.mlp.up_proj.weight": [ 6144, 2048 ], "thinker.model.language_model.layers.21.mlp.down_proj.weight": [ 2048, 6144 ], "thinker.model.language_model.layers.21.input_layernorm.weight": [ 2048 ], "thinker.model.language_model.layers.21.post_attention_layernorm.weight": [ 2048 ], "thinker.model.language_model.layers.22.linear_attn.dt_bias": [ 16 ], "thinker.model.language_model.layers.22.linear_attn.A_log": [ 16 ], "thinker.model.language_model.layers.22.linear_attn.conv1d.weight": [ 6144, 1, 4 ], "thinker.model.language_model.layers.22.linear_attn.norm.weight": [ 128 ], "thinker.model.language_model.layers.22.linear_attn.out_proj.weight": [ 2048, 2048 ], "thinker.model.language_model.layers.22.linear_attn.in_proj_qkv.weight": [ 6144, 2048 ], "thinker.model.language_model.layers.22.linear_attn.in_proj_z.weight": [ 2048, 2048 ], "thinker.model.language_model.layers.22.linear_attn.in_proj_b.weight": [ 16, 2048 ], "thinker.model.language_model.layers.22.linear_attn.in_proj_a.weight": [ 16, 2048 ], "thinker.model.language_model.layers.22.mlp.gate_proj.weight": [ 6144, 2048 ], "thinker.model.language_model.layers.22.mlp.up_proj.weight": [ 6144, 2048 ], "thinker.model.language_model.layers.22.mlp.down_proj.weight": [ 2048, 6144 ], "thinker.model.language_model.layers.22.input_layernorm.weight": [ 2048 ], "thinker.model.language_model.layers.22.post_attention_layernorm.weight": [ 2048 ], "thinker.model.language_model.layers.23.self_attn.q_proj.weight": [ 4096, 2048 ], "thinker.model.language_model.layers.23.self_attn.k_proj.weight": [ 512, 2048 ], "thinker.model.language_model.layers.23.self_attn.v_proj.weight": [ 512, 2048 ], "thinker.model.language_model.layers.23.self_attn.o_proj.weight": [ 2048, 2048 ], "thinker.model.language_model.layers.23.self_attn.q_norm.weight": [ 256 ], "thinker.model.language_model.layers.23.self_attn.k_norm.weight": [ 256 ], "thinker.model.language_model.layers.23.mlp.gate_proj.weight": [ 6144, 2048 ], "thinker.model.language_model.layers.23.mlp.up_proj.weight": [ 6144, 2048 ], "thinker.model.language_model.layers.23.mlp.down_proj.weight": [ 2048, 6144 ], "thinker.model.language_model.layers.23.input_layernorm.weight": [ 2048 ], "thinker.model.language_model.layers.23.post_attention_layernorm.weight": [ 2048 ], "thinker.model.language_model.norm.weight": [ 2048 ], "audio_tower.layers.22.self_attn.k_proj.weight": [ 1024, 1024 ], "audio_tower.layers.22.self_attn.k_proj.bias": [ 1024 ], "audio_tower.layers.22.self_attn.v_proj.weight": [ 1024, 1024 ], "audio_tower.layers.22.self_attn.v_proj.bias": [ 1024 ], "audio_tower.layers.22.self_attn.q_proj.weight": [ 1024, 1024 ], "audio_tower.layers.22.self_attn.q_proj.bias": [ 1024 ], "audio_tower.layers.22.self_attn.out_proj.weight": [ 1024, 1024 ], "audio_tower.layers.22.self_attn.out_proj.bias": [ 1024 ], "audio_tower.layers.22.self_attn_layer_norm.weight": [ 1024 ], "audio_tower.layers.22.self_attn_layer_norm.bias": [ 1024 ], "audio_tower.layers.22.fc1.weight": [ 4096, 1024 ], "audio_tower.layers.22.fc1.bias": [ 4096 ], "audio_tower.layers.22.fc2.weight": [ 1024, 4096 ], "audio_tower.layers.22.fc2.bias": [ 1024 ], "audio_tower.layers.22.final_layer_norm.weight": [ 1024 ], "audio_tower.layers.22.final_layer_norm.bias": [ 1024 ], "audio_tower.layers.23.self_attn.k_proj.weight": [ 1024, 1024 ], "audio_tower.layers.23.self_attn.k_proj.bias": [ 1024 ], "audio_tower.layers.23.self_attn.v_proj.weight": [ 1024, 1024 ], "audio_tower.layers.23.self_attn.v_proj.bias": [ 1024 ], "audio_tower.layers.23.self_attn.q_proj.weight": [ 1024, 1024 ], "audio_tower.layers.23.self_attn.q_proj.bias": [ 1024 ], "audio_tower.layers.23.self_attn.out_proj.weight": [ 1024, 1024 ], "audio_tower.layers.23.self_attn.out_proj.bias": [ 1024 ], "audio_tower.layers.23.self_attn_layer_norm.weight": [ 1024 ], "audio_tower.layers.23.self_attn_layer_norm.bias": [ 1024 ], "audio_tower.layers.23.fc1.weight": [ 4096, 1024 ], "audio_tower.layers.23.fc1.bias": [ 4096 ], "audio_tower.layers.23.fc2.weight": [ 1024, 4096 ], "audio_tower.layers.23.fc2.bias": [ 1024 ], "audio_tower.layers.23.final_layer_norm.weight": [ 1024 ], "audio_tower.layers.23.final_layer_norm.bias": [ 1024 ], "audio_tower.ln_post.weight": [ 1024 ], "audio_tower.ln_post.bias": [ 1024 ], "audio_tower.proj1.weight": [ 1024, 1024 ], "audio_tower.proj1.bias": [ 1024 ], "audio_tower.proj2.weight": [ 2048, 1024 ], "audio_tower.proj2.bias": [ 2048 ], "align.gate": [], "align.out_scale": [], "align.norm.weight": [ 2048 ], "align.norm.bias": [ 2048 ], "align.proj.weight": [ 16384, 2048 ], "align.proj.bias": [ 16384 ], "align.proj2.weight": [ 2048, 16384 ], "align.proj2.bias": [ 2048 ], "audio_special.weight": [ 3, 2048 ] }