<
   ir_version: 10,
   opset_import: ["" : 23],
   producer_name: "pytorch"
>
main_graph (int32[batch,64] text) => (float[batch,1152] text_embedding) 
   <
      float[batch,64,1152] add_1071
      float[batch,64,1152] add_1092
      float[batch,64,1152] add_119
      float[batch,64,1152] add_1207
      float[batch,64,1152] add_1228
      float[batch,64,1152] add_1343
      float[batch,64,1152] add_1364
      float[batch,64,1152] add_140
      float[batch,64,1152] add_1479
      float[batch,64,1152] add_1500
      float[batch,64,1152] add_1615
      float[batch,64,1152] add_1636
      float[batch,64,1152] add_1751
      float[batch,64,1152] add_1772
      float[batch,64,1152] add_1887
      float[batch,64,1152] add_1908
      float[batch,64,1152] add_2023
      float[batch,64,1152] add_2044
      float[batch,64,1152] add_2159
      float[batch,64,1152] add_2180
      float[batch,64,1152] add_2295
      float[batch,64,1152] add_2316
      float[batch,64,1152] add_2431
      float[batch,64,1152] add_2452
      float[batch,64,1152] add_255
      float[batch,64,1152] add_2567
      float[batch,64,1152] add_2588
      float[batch,64,1152] add_2703
      float[batch,64,1152] add_2724
      float[batch,64,1152] add_276
      float[batch,64,1152] add_2839
      float[batch,64,1152] add_2860
      float[batch,64,1152] add_2975
      float[batch,64,1152] add_2996
      float[batch,64,1152] add_3111
      float[batch,64,1152] add_3132
      float[batch,64,1152] add_3247
      float[batch,64,1152] add_3268
      float[batch,64,1152] add_3383
      float[batch,64,1152] add_3404
      float[batch,64,1152] add_3519
      float[batch,64,1152] add_3540
      float[batch,1,1152] add_3540_pooled
      float[batch,1,1152] add_3655
      float[batch,1,1152] add_3676
      float[batch,64,1152] add_391
      float[batch,64,1152] add_4
      float[batch,64,1152] add_412
      float[batch,64,1152] add_527
      float[batch,64,1152] add_548
      float[batch,64,1152] add_663
      float[batch,64,1152] add_684
      float[batch,64,1152] add_799
      float[batch,64,1152] add_820
      float[batch,64,1152] add_935
      float[batch,64,1152] add_956
      float[batch,1] clamp_min
      float[batch,64,1152] embedding
      float[batch,64,4304] gelu
      float[batch,64,4304] gelu_1
      float[batch,64,4304] gelu_10
      float[batch,64,4304] gelu_11
      float[batch,64,4304] gelu_12
      float[batch,64,4304] gelu_13
      float[batch,64,4304] gelu_14
      float[batch,64,4304] gelu_15
      float[batch,64,4304] gelu_16
      float[batch,64,4304] gelu_17
      float[batch,64,4304] gelu_18
      float[batch,64,4304] gelu_19
      float[batch,64,4304] gelu_2
      float[batch,64,4304] gelu_20
      float[batch,64,4304] gelu_21
      float[batch,64,4304] gelu_22
      float[batch,64,4304] gelu_23
      float[batch,64,4304] gelu_24
      float[batch,64,4304] gelu_25
      float[batch,1,4304] gelu_26
      float[batch,64,4304] gelu_3
      float[batch,64,4304] gelu_4
      float[batch,64,4304] gelu_5
      float[batch,64,4304] gelu_6
      float[batch,64,4304] gelu_7
      float[batch,64,4304] gelu_8
      float[batch,64,4304] gelu_9
      float[batch,64,1152] layer_norm
      float[batch,64,1152] layer_norm_1
      float[batch,64,1152] layer_norm_10
      float[batch,64,1152] layer_norm_11
      float[batch,64,1152] layer_norm_12
      float[batch,64,1152] layer_norm_13
      float[batch,64,1152] layer_norm_14
      float[batch,64,1152] layer_norm_15
      float[batch,64,1152] layer_norm_16
      float[batch,64,1152] layer_norm_17
      float[batch,64,1152] layer_norm_18
      float[batch,64,1152] layer_norm_19
      float[batch,64,1152] layer_norm_2
      float[batch,64,1152] layer_norm_20
      float[batch,64,1152] layer_norm_21
      float[batch,64,1152] layer_norm_22
      float[batch,64,1152] layer_norm_23
      float[batch,64,1152] layer_norm_24
      float[batch,64,1152] layer_norm_25
      float[batch,64,1152] layer_norm_26
      float[batch,64,1152] layer_norm_27
      float[batch,64,1152] layer_norm_28
      float[batch,64,1152] layer_norm_29
      float[batch,64,1152] layer_norm_3
      float[batch,64,1152] layer_norm_30
      float[batch,64,1152] layer_norm_31
      float[batch,64,1152] layer_norm_32
      float[batch,64,1152] layer_norm_33
      float[batch,64,1152] layer_norm_34
      float[batch,64,1152] layer_norm_35
      float[batch,64,1152] layer_norm_36
      float[batch,64,1152] layer_norm_37
      float[batch,64,1152] layer_norm_38
      float[batch,64,1152] layer_norm_39
      float[batch,64,1152] layer_norm_4
      float[batch,64,1152] layer_norm_40
      float[batch,64,1152] layer_norm_41
      float[batch,64,1152] layer_norm_42
      float[batch,64,1152] layer_norm_43
      float[batch,64,1152] layer_norm_44
      float[batch,64,1152] layer_norm_45
      float[batch,64,1152] layer_norm_46
      float[batch,64,1152] layer_norm_47
      float[batch,64,1152] layer_norm_48
      float[batch,64,1152] layer_norm_49
      float[batch,64,1152] layer_norm_5
      float[batch,64,1152] layer_norm_50
      float[batch,64,1152] layer_norm_51
      float[batch,64,1152] layer_norm_52
      float[batch,1,1152] layer_norm_53
      float[batch,64,1152] layer_norm_6
      float[batch,64,1152] layer_norm_7
      float[batch,64,1152] layer_norm_8
      float[batch,64,1152] layer_norm_9
      float[batch,1] linalg_vector_norm
      float[batch,64,4304] linear_10
      float[batch,64,4304] linear_102
      float[batch,64,1152] linear_103
      float[batch,1,4304] linear_106
      float[batch,1,1152] linear_107
      float[batch,1152] linear_108
      float[batch,64,1152] linear_11
      float[batch,64,4304] linear_14
      float[batch,64,1152] linear_15
      float[batch,64,4304] linear_18
      float[batch,64,1152] linear_19
      float[batch,64,4304] linear_2
      float[batch,64,4304] linear_22
      float[batch,64,1152] linear_23
      float[batch,64,4304] linear_26
      float[batch,64,1152] linear_27
      float[batch,64,1152] linear_3
      float[batch,64,4304] linear_30
      float[batch,64,1152] linear_31
      float[batch,64,4304] linear_34
      float[batch,64,1152] linear_35
      float[batch,64,4304] linear_38
      float[batch,64,1152] linear_39
      float[batch,64,4304] linear_42
      float[batch,64,1152] linear_43
      float[batch,64,4304] linear_46
      float[batch,64,1152] linear_47
      float[batch,64,4304] linear_50
      float[batch,64,1152] linear_51
      float[batch,64,4304] linear_54
      float[batch,64,1152] linear_55
      float[batch,64,4304] linear_58
      float[batch,64,1152] linear_59
      float[batch,64,4304] linear_6
      float[batch,64,4304] linear_62
      float[batch,64,1152] linear_63
      float[batch,64,4304] linear_66
      float[batch,64,1152] linear_67
      float[batch,64,1152] linear_7
      float[batch,64,4304] linear_70
      float[batch,64,1152] linear_71
      float[batch,64,4304] linear_74
      float[batch,64,1152] linear_75
      float[batch,64,4304] linear_78
      float[batch,64,1152] linear_79
      float[batch,64,4304] linear_82
      float[batch,64,1152] linear_83
      float[batch,64,4304] linear_86
      float[batch,64,1152] linear_87
      float[batch,64,4304] linear_90
      float[batch,64,1152] linear_91
      float[batch,64,4304] linear_94
      float[batch,64,1152] linear_95
      float[batch,64,4304] linear_98
      float[batch,64,1152] linear_99
      float[batch,64,1152] node_scaled_dot_product_attention_10_k
      float[batch,64,1152] node_scaled_dot_product_attention_10_out
      float[batch,64,1152] node_scaled_dot_product_attention_10_out_mm_out
      float[batch,64,1152] node_scaled_dot_product_attention_10_q
      float[batch,64,3456] node_scaled_dot_product_attention_10_qkv
      float[batch,64,3456] node_scaled_dot_product_attention_10_qkv_mm_out
      float[batch,64,1152] node_scaled_dot_product_attention_10_v
      float[batch,64,1152] node_scaled_dot_product_attention_11_k
      float[batch,64,1152] node_scaled_dot_product_attention_11_out
      float[batch,64,1152] node_scaled_dot_product_attention_11_out_mm_out
      float[batch,64,1152] node_scaled_dot_product_attention_11_q
      float[batch,64,3456] node_scaled_dot_product_attention_11_qkv
      float[batch,64,3456] node_scaled_dot_product_attention_11_qkv_mm_out
      float[batch,64,1152] node_scaled_dot_product_attention_11_v
      float[batch,64,1152] node_scaled_dot_product_attention_12_k
      float[batch,64,1152] node_scaled_dot_product_attention_12_out
      float[batch,64,1152] node_scaled_dot_product_attention_12_out_mm_out
      float[batch,64,1152] node_scaled_dot_product_attention_12_q
      float[batch,64,3456] node_scaled_dot_product_attention_12_qkv
      float[batch,64,3456] node_scaled_dot_product_attention_12_qkv_mm_out
      float[batch,64,1152] node_scaled_dot_product_attention_12_v
      float[batch,64,1152] node_scaled_dot_product_attention_13_k
      float[batch,64,1152] node_scaled_dot_product_attention_13_out
      float[batch,64,1152] node_scaled_dot_product_attention_13_out_mm_out
      float[batch,64,1152] node_scaled_dot_product_attention_13_q
      float[batch,64,3456] node_scaled_dot_product_attention_13_qkv
      float[batch,64,3456] node_scaled_dot_product_attention_13_qkv_mm_out
      float[batch,64,1152] node_scaled_dot_product_attention_13_v
      float[batch,64,1152] node_scaled_dot_product_attention_14_k
      float[batch,64,1152] node_scaled_dot_product_attention_14_out
      float[batch,64,1152] node_scaled_dot_product_attention_14_out_mm_out
      float[batch,64,1152] node_scaled_dot_product_attention_14_q
      float[batch,64,3456] node_scaled_dot_product_attention_14_qkv
      float[batch,64,3456] node_scaled_dot_product_attention_14_qkv_mm_out
      float[batch,64,1152] node_scaled_dot_product_attention_14_v
      float[batch,64,1152] node_scaled_dot_product_attention_15_k
      float[batch,64,1152] node_scaled_dot_product_attention_15_out
      float[batch,64,1152] node_scaled_dot_product_attention_15_out_mm_out
      float[batch,64,1152] node_scaled_dot_product_attention_15_q
      float[batch,64,3456] node_scaled_dot_product_attention_15_qkv
      float[batch,64,3456] node_scaled_dot_product_attention_15_qkv_mm_out
      float[batch,64,1152] node_scaled_dot_product_attention_15_v
      float[batch,64,1152] node_scaled_dot_product_attention_16_k
      float[batch,64,1152] node_scaled_dot_product_attention_16_out
      float[batch,64,1152] node_scaled_dot_product_attention_16_out_mm_out
      float[batch,64,1152] node_scaled_dot_product_attention_16_q
      float[batch,64,3456] node_scaled_dot_product_attention_16_qkv
      float[batch,64,3456] node_scaled_dot_product_attention_16_qkv_mm_out
      float[batch,64,1152] node_scaled_dot_product_attention_16_v
      float[batch,64,1152] node_scaled_dot_product_attention_17_k
      float[batch,64,1152] node_scaled_dot_product_attention_17_out
      float[batch,64,1152] node_scaled_dot_product_attention_17_out_mm_out
      float[batch,64,1152] node_scaled_dot_product_attention_17_q
      float[batch,64,3456] node_scaled_dot_product_attention_17_qkv
      float[batch,64,3456] node_scaled_dot_product_attention_17_qkv_mm_out
      float[batch,64,1152] node_scaled_dot_product_attention_17_v
      float[batch,64,1152] node_scaled_dot_product_attention_18_k
      float[batch,64,1152] node_scaled_dot_product_attention_18_out
      float[batch,64,1152] node_scaled_dot_product_attention_18_out_mm_out
      float[batch,64,1152] node_scaled_dot_product_attention_18_q
      float[batch,64,3456] node_scaled_dot_product_attention_18_qkv
      float[batch,64,3456] node_scaled_dot_product_attention_18_qkv_mm_out
      float[batch,64,1152] node_scaled_dot_product_attention_18_v
      float[batch,64,1152] node_scaled_dot_product_attention_19_k
      float[batch,64,1152] node_scaled_dot_product_attention_19_out
      float[batch,64,1152] node_scaled_dot_product_attention_19_out_mm_out
      float[batch,64,1152] node_scaled_dot_product_attention_19_q
      float[batch,64,3456] node_scaled_dot_product_attention_19_qkv
      float[batch,64,3456] node_scaled_dot_product_attention_19_qkv_mm_out
      float[batch,64,1152] node_scaled_dot_product_attention_19_v
      float[batch,64,1152] node_scaled_dot_product_attention_1_k
      float[batch,64,1152] node_scaled_dot_product_attention_1_out
      float[batch,64,1152] node_scaled_dot_product_attention_1_out_mm_out
      float[batch,64,1152] node_scaled_dot_product_attention_1_q
      float[batch,64,3456] node_scaled_dot_product_attention_1_qkv
      float[batch,64,3456] node_scaled_dot_product_attention_1_qkv_mm_out
      float[batch,64,1152] node_scaled_dot_product_attention_1_v
      float[batch,64,1152] node_scaled_dot_product_attention_20_k
      float[batch,64,1152] node_scaled_dot_product_attention_20_out
      float[batch,64,1152] node_scaled_dot_product_attention_20_out_mm_out
      float[batch,64,1152] node_scaled_dot_product_attention_20_q
      float[batch,64,3456] node_scaled_dot_product_attention_20_qkv
      float[batch,64,3456] node_scaled_dot_product_attention_20_qkv_mm_out
      float[batch,64,1152] node_scaled_dot_product_attention_20_v
      float[batch,64,1152] node_scaled_dot_product_attention_21_k
      float[batch,64,1152] node_scaled_dot_product_attention_21_out
      float[batch,64,1152] node_scaled_dot_product_attention_21_out_mm_out
      float[batch,64,1152] node_scaled_dot_product_attention_21_q
      float[batch,64,3456] node_scaled_dot_product_attention_21_qkv
      float[batch,64,3456] node_scaled_dot_product_attention_21_qkv_mm_out
      float[batch,64,1152] node_scaled_dot_product_attention_21_v
      float[batch,64,1152] node_scaled_dot_product_attention_22_k
      float[batch,64,1152] node_scaled_dot_product_attention_22_out
      float[batch,64,1152] node_scaled_dot_product_attention_22_out_mm_out
      float[batch,64,1152] node_scaled_dot_product_attention_22_q
      float[batch,64,3456] node_scaled_dot_product_attention_22_qkv
      float[batch,64,3456] node_scaled_dot_product_attention_22_qkv_mm_out
      float[batch,64,1152] node_scaled_dot_product_attention_22_v
      float[batch,64,1152] node_scaled_dot_product_attention_23_k
      float[batch,64,1152] node_scaled_dot_product_attention_23_out
      float[batch,64,1152] node_scaled_dot_product_attention_23_out_mm_out
      float[batch,64,1152] node_scaled_dot_product_attention_23_q
      float[batch,64,3456] node_scaled_dot_product_attention_23_qkv
      float[batch,64,3456] node_scaled_dot_product_attention_23_qkv_mm_out
      float[batch,64,1152] node_scaled_dot_product_attention_23_v
      float[batch,64,1152] node_scaled_dot_product_attention_24_k
      float[batch,64,1152] node_scaled_dot_product_attention_24_out
      float[batch,64,1152] node_scaled_dot_product_attention_24_out_mm_out
      float[batch,64,1152] node_scaled_dot_product_attention_24_q
      float[batch,64,3456] node_scaled_dot_product_attention_24_qkv
      float[batch,64,3456] node_scaled_dot_product_attention_24_qkv_mm_out
      float[batch,64,1152] node_scaled_dot_product_attention_24_v
      float[batch,64,1152] node_scaled_dot_product_attention_25_k
      float[batch,64,1152] node_scaled_dot_product_attention_25_out
      float[batch,64,1152] node_scaled_dot_product_attention_25_out_mm_out
      float[batch,64,1152] node_scaled_dot_product_attention_25_q
      float[batch,64,3456] node_scaled_dot_product_attention_25_qkv
      float[batch,64,3456] node_scaled_dot_product_attention_25_qkv_mm_out
      float[batch,64,1152] node_scaled_dot_product_attention_25_v
      float[batch,64,1152] node_scaled_dot_product_attention_26_k
      float[batch,1,1152] node_scaled_dot_product_attention_26_out
      float[batch,1,1152] node_scaled_dot_product_attention_26_out_mm_out
      float[batch,64,1152] node_scaled_dot_product_attention_26_q
      float[batch,1,1152] node_scaled_dot_product_attention_26_q_pooled
      float[batch,64,3456] node_scaled_dot_product_attention_26_qkv
      float[batch,64,3456] node_scaled_dot_product_attention_26_qkv_mm_out
      float[batch,64,1152] node_scaled_dot_product_attention_26_v
      float[batch,64,1152] node_scaled_dot_product_attention_2_k
      float[batch,64,1152] node_scaled_dot_product_attention_2_out
      float[batch,64,1152] node_scaled_dot_product_attention_2_out_mm_out
      float[batch,64,1152] node_scaled_dot_product_attention_2_q
      float[batch,64,3456] node_scaled_dot_product_attention_2_qkv
      float[batch,64,3456] node_scaled_dot_product_attention_2_qkv_mm_out
      float[batch,64,1152] node_scaled_dot_product_attention_2_v
      float[batch,64,1152] node_scaled_dot_product_attention_3_k
      float[batch,64,1152] node_scaled_dot_product_attention_3_out
      float[batch,64,1152] node_scaled_dot_product_attention_3_out_mm_out
      float[batch,64,1152] node_scaled_dot_product_attention_3_q
      float[batch,64,3456] node_scaled_dot_product_attention_3_qkv
      float[batch,64,3456] node_scaled_dot_product_attention_3_qkv_mm_out
      float[batch,64,1152] node_scaled_dot_product_attention_3_v
      float[batch,64,1152] node_scaled_dot_product_attention_4_k
      float[batch,64,1152] node_scaled_dot_product_attention_4_out
      float[batch,64,1152] node_scaled_dot_product_attention_4_out_mm_out
      float[batch,64,1152] node_scaled_dot_product_attention_4_q
      float[batch,64,3456] node_scaled_dot_product_attention_4_qkv
      float[batch,64,3456] node_scaled_dot_product_attention_4_qkv_mm_out
      float[batch,64,1152] node_scaled_dot_product_attention_4_v
      float[batch,64,1152] node_scaled_dot_product_attention_5_k
      float[batch,64,1152] node_scaled_dot_product_attention_5_out
      float[batch,64,1152] node_scaled_dot_product_attention_5_out_mm_out
      float[batch,64,1152] node_scaled_dot_product_attention_5_q
      float[batch,64,3456] node_scaled_dot_product_attention_5_qkv
      float[batch,64,3456] node_scaled_dot_product_attention_5_qkv_mm_out
      float[batch,64,1152] node_scaled_dot_product_attention_5_v
      float[batch,64,1152] node_scaled_dot_product_attention_6_k
      float[batch,64,1152] node_scaled_dot_product_attention_6_out
      float[batch,64,1152] node_scaled_dot_product_attention_6_out_mm_out
      float[batch,64,1152] node_scaled_dot_product_attention_6_q
      float[batch,64,3456] node_scaled_dot_product_attention_6_qkv
      float[batch,64,3456] node_scaled_dot_product_attention_6_qkv_mm_out
      float[batch,64,1152] node_scaled_dot_product_attention_6_v
      float[batch,64,1152] node_scaled_dot_product_attention_7_k
      float[batch,64,1152] node_scaled_dot_product_attention_7_out
      float[batch,64,1152] node_scaled_dot_product_attention_7_out_mm_out
      float[batch,64,1152] node_scaled_dot_product_attention_7_q
      float[batch,64,3456] node_scaled_dot_product_attention_7_qkv
      float[batch,64,3456] node_scaled_dot_product_attention_7_qkv_mm_out
      float[batch,64,1152] node_scaled_dot_product_attention_7_v
      float[batch,64,1152] node_scaled_dot_product_attention_8_k
      float[batch,64,1152] node_scaled_dot_product_attention_8_out
      float[batch,64,1152] node_scaled_dot_product_attention_8_out_mm_out
      float[batch,64,1152] node_scaled_dot_product_attention_8_q
      float[batch,64,3456] node_scaled_dot_product_attention_8_qkv
      float[batch,64,3456] node_scaled_dot_product_attention_8_qkv_mm_out
      float[batch,64,1152] node_scaled_dot_product_attention_8_v
      float[batch,64,1152] node_scaled_dot_product_attention_9_k
      float[batch,64,1152] node_scaled_dot_product_attention_9_out
      float[batch,64,1152] node_scaled_dot_product_attention_9_out_mm_out
      float[batch,64,1152] node_scaled_dot_product_attention_9_q
      float[batch,64,3456] node_scaled_dot_product_attention_9_qkv
      float[batch,64,3456] node_scaled_dot_product_attention_9_qkv_mm_out
      float[batch,64,1152] node_scaled_dot_product_attention_9_v
      float[batch,64,1152] node_scaled_dot_product_attention_k
      float[batch,64,1152] node_scaled_dot_product_attention_out
      float[batch,64,1152] node_scaled_dot_product_attention_out_mm_out
      float[batch,64,1152] node_scaled_dot_product_attention_q
      float[batch,64,3456] node_scaled_dot_product_attention_qkv
      float[batch,64,3456] node_scaled_dot_product_attention_qkv_mm_out
      float[batch,64,1152] node_scaled_dot_product_attention_v
      float[batch,64,1152] scaled_dot_product_attention
      float[batch,64,1152] scaled_dot_product_attention_1
      float[batch,64,1152] scaled_dot_product_attention_10
      float[batch,64,1152] scaled_dot_product_attention_11
      float[batch,64,1152] scaled_dot_product_attention_12
      float[batch,64,1152] scaled_dot_product_attention_13
      float[batch,64,1152] scaled_dot_product_attention_14
      float[batch,64,1152] scaled_dot_product_attention_15
      float[batch,64,1152] scaled_dot_product_attention_16
      float[batch,64,1152] scaled_dot_product_attention_17
      float[batch,64,1152] scaled_dot_product_attention_18
      float[batch,64,1152] scaled_dot_product_attention_19
      float[batch,64,1152] scaled_dot_product_attention_2
      float[batch,64,1152] scaled_dot_product_attention_20
      float[batch,64,1152] scaled_dot_product_attention_21
      float[batch,64,1152] scaled_dot_product_attention_22
      float[batch,64,1152] scaled_dot_product_attention_23
      float[batch,64,1152] scaled_dot_product_attention_24
      float[batch,64,1152] scaled_dot_product_attention_25
      float[batch,1,1152] scaled_dot_product_attention_26
      float[batch,64,1152] scaled_dot_product_attention_3
      float[batch,64,1152] scaled_dot_product_attention_4
      float[batch,64,1152] scaled_dot_product_attention_5
      float[batch,64,1152] scaled_dot_product_attention_6
      float[batch,64,1152] scaled_dot_product_attention_7
      float[batch,64,1152] scaled_dot_product_attention_8
      float[batch,64,1152] scaled_dot_product_attention_9
      float[batch,1152] select_81
      float[batch,64,4304] val_100
      float[batch,64,1152] val_101
      float[batch,64,4304] val_102
      float[batch,64,1152] val_103
      float[batch,64,4304] val_104
      float[batch,64,1152] val_105
      float[batch,64,4304] val_106
      float[batch,64,1152] val_107
      float[batch,64,4304] val_108
      float[batch,64,1152] val_109
      float[batch,64,4304] val_110
      float[batch,64,1152] val_111
      float[batch,64,4304] val_112
      float[batch,64,1152] val_113
      float[batch,64,4304] val_114
      float[batch,64,1152] val_115
      float[batch,64,4304] val_116
      float[batch,64,1152] val_117
      float[batch,64,4304] val_118
      float[batch,64,1152] val_119
      float[batch,64,4304] val_120
      float[batch,64,1152] val_121
      float[batch,64,4304] val_122
      float[batch,64,1152] val_123
      float[batch,64,4304] val_124
      float[batch,64,1152] val_125
      float[batch,64,4304] val_126
      float[batch,64,1152] val_127
      float[batch,64,4304] val_128
      float[batch,64,1152] val_129
      float[batch,64,4304] val_130
      float[batch,64,1152] val_131
      float[batch,64,4304] val_132
      float[batch,64,1152] val_133
      float[batch,64,4304] val_134
      float[batch,64,1152] val_135
      float[batch,64,4304] val_136
      float[batch,64,1152] val_137
      float[batch,1,4304] val_138
      float[batch,1,1152] val_139
      float[batch,1152] val_140
      float[batch,64,4304] val_86
      float[batch,64,1152] val_87
      float[batch,64,4304] val_88
      float[batch,64,1152] val_89
      float[batch,64,4304] val_90
      float[batch,64,1152] val_91
      float[batch,64,4304] val_92
      float[batch,64,1152] val_93
      float[batch,64,4304] val_94
      float[batch,64,1152] val_95
      float[batch,64,4304] val_96
      float[batch,64,1152] val_97
      float[batch,64,4304] val_98
      float[batch,64,1152] val_99
   >
{
   val_85 = Gather <axis: int = 0> ("text.token_embedding.weight_fp16", text)
   embedding = Cast <to: int = 1> (val_85)
   add_4 = Add (embedding, "text.positional_embedding")
   [node_layer_norm] layer_norm = LayerNormalization <axis: int = -1, epsilon: float = 1e-06, stash_type: int = 1> (add_4, "text.transformer.resblocks.0.ln_1.weight", "text.transformer.resblocks.0.ln_1.bias")
   [node_scaled_dot_product_attention_qkv_mm] node_scaled_dot_product_attention_qkv_mm_out = MatMul (layer_norm, val_4)
   [node_scaled_dot_product_attention_qkv_bias] node_scaled_dot_product_attention_qkv = Add (node_scaled_dot_product_attention_qkv_mm_out, "text.transformer.resblocks.0.attn.in_proj_bias")
   [node_scaled_dot_product_attention_qkv_split] node_scaled_dot_product_attention_q, node_scaled_dot_product_attention_k, node_scaled_dot_product_attention_v = Split <axis: int = -1> (node_scaled_dot_product_attention_qkv, attn3d_split_3x1152)
   [node_scaled_dot_product_attention] scaled_dot_product_attention = Attention <is_causal: int = 0, kv_num_heads: int = 16, q_num_heads: int = 16, qk_matmul_output_mode: int = 0, softcap: float = 0> (node_scaled_dot_product_attention_q, node_scaled_dot_product_attention_k, node_scaled_dot_product_attention_v)
   [node_scaled_dot_product_attention_out_mm] node_scaled_dot_product_attention_out_mm_out = MatMul (scaled_dot_product_attention, node_scaled_dot_product_attention_wo_t)
   [node_scaled_dot_product_attention_out_bias] node_scaled_dot_product_attention_out = Add (node_scaled_dot_product_attention_out_mm_out, "text.transformer.resblocks.0.attn.out_proj.bias")
   add_119 = Add (add_4, node_scaled_dot_product_attention_out)
   layer_norm_1 = LayerNormalization <axis: int = -1, epsilon: float = 1e-06, stash_type: int = 1> (add_119, "text.transformer.resblocks.0.ln_2.weight", "text.transformer.resblocks.0.ln_2.bias")
   val_86 = MatMul (layer_norm_1, val_5)
   linear_2 = Add (val_86, "text.transformer.resblocks.0.mlp.c_fc.bias")
   [node_gelu] gelu = Gelu <approximate: string = "tanh"> (linear_2)
   val_87 = MatMul (gelu, val_6)
   linear_3 = Add (val_87, "text.transformer.resblocks.0.mlp.c_proj.bias")
   add_140 = Add (add_119, linear_3)
   layer_norm_2 = LayerNormalization <axis: int = -1, epsilon: float = 1e-06, stash_type: int = 1> (add_140, "text.transformer.resblocks.1.ln_1.weight", "text.transformer.resblocks.1.ln_1.bias")
   [node_scaled_dot_product_attention_1_qkv_mm] node_scaled_dot_product_attention_1_qkv_mm_out = MatMul (layer_norm_2, val_7)
   [node_scaled_dot_product_attention_1_qkv_bias] node_scaled_dot_product_attention_1_qkv = Add (node_scaled_dot_product_attention_1_qkv_mm_out, "text.transformer.resblocks.1.attn.in_proj_bias")
   [node_scaled_dot_product_attention_1_qkv_split] node_scaled_dot_product_attention_1_q, node_scaled_dot_product_attention_1_k, node_scaled_dot_product_attention_1_v = Split <axis: int = -1> (node_scaled_dot_product_attention_1_qkv, attn3d_split_3x1152)
   scaled_dot_product_attention_1 = Attention <is_causal: int = 0, kv_num_heads: int = 16, q_num_heads: int = 16, qk_matmul_output_mode: int = 0, softcap: float = 0> (node_scaled_dot_product_attention_1_q, node_scaled_dot_product_attention_1_k, node_scaled_dot_product_attention_1_v)
   [node_scaled_dot_product_attention_1_out_mm] node_scaled_dot_product_attention_1_out_mm_out = MatMul (scaled_dot_product_attention_1, node_scaled_dot_product_attention_1_wo_t)
   [node_scaled_dot_product_attention_1_out_bias] node_scaled_dot_product_attention_1_out = Add (node_scaled_dot_product_attention_1_out_mm_out, "text.transformer.resblocks.1.attn.out_proj.bias")
   add_255 = Add (add_140, node_scaled_dot_product_attention_1_out)
   layer_norm_3 = LayerNormalization <axis: int = -1, epsilon: float = 1e-06, stash_type: int = 1> (add_255, "text.transformer.resblocks.1.ln_2.weight", "text.transformer.resblocks.1.ln_2.bias")
   val_88 = MatMul (layer_norm_3, val_8)
   linear_6 = Add (val_88, "text.transformer.resblocks.1.mlp.c_fc.bias")
   gelu_1 = Gelu <approximate: string = "tanh"> (linear_6)
   val_89 = MatMul (gelu_1, val_9)
   linear_7 = Add (val_89, "text.transformer.resblocks.1.mlp.c_proj.bias")
   add_276 = Add (add_255, linear_7)
   layer_norm_4 = LayerNormalization <axis: int = -1, epsilon: float = 1e-06, stash_type: int = 1> (add_276, "text.transformer.resblocks.2.ln_1.weight", "text.transformer.resblocks.2.ln_1.bias")
   [node_scaled_dot_product_attention_2_qkv_mm] node_scaled_dot_product_attention_2_qkv_mm_out = MatMul (layer_norm_4, val_10)
   [node_scaled_dot_product_attention_2_qkv_bias] node_scaled_dot_product_attention_2_qkv = Add (node_scaled_dot_product_attention_2_qkv_mm_out, "text.transformer.resblocks.2.attn.in_proj_bias")
   [node_scaled_dot_product_attention_2_qkv_split] node_scaled_dot_product_attention_2_q, node_scaled_dot_product_attention_2_k, node_scaled_dot_product_attention_2_v = Split <axis: int = -1> (node_scaled_dot_product_attention_2_qkv, attn3d_split_3x1152)
   scaled_dot_product_attention_2 = Attention <is_causal: int = 0, kv_num_heads: int = 16, q_num_heads: int = 16, qk_matmul_output_mode: int = 0, softcap: float = 0> (node_scaled_dot_product_attention_2_q, node_scaled_dot_product_attention_2_k, node_scaled_dot_product_attention_2_v)
   [node_scaled_dot_product_attention_2_out_mm] node_scaled_dot_product_attention_2_out_mm_out = MatMul (scaled_dot_product_attention_2, node_scaled_dot_product_attention_2_wo_t)
   [node_scaled_dot_product_attention_2_out_bias] node_scaled_dot_product_attention_2_out = Add (node_scaled_dot_product_attention_2_out_mm_out, "text.transformer.resblocks.2.attn.out_proj.bias")
   add_391 = Add (add_276, node_scaled_dot_product_attention_2_out)
   layer_norm_5 = LayerNormalization <axis: int = -1, epsilon: float = 1e-06, stash_type: int = 1> (add_391, "text.transformer.resblocks.2.ln_2.weight", "text.transformer.resblocks.2.ln_2.bias")
   val_90 = MatMul (layer_norm_5, val_11)
   linear_10 = Add (val_90, "text.transformer.resblocks.2.mlp.c_fc.bias")
   gelu_2 = Gelu <approximate: string = "tanh"> (linear_10)
   val_91 = MatMul (gelu_2, val_12)
   linear_11 = Add (val_91, "text.transformer.resblocks.2.mlp.c_proj.bias")
   add_412 = Add (add_391, linear_11)
   layer_norm_6 = LayerNormalization <axis: int = -1, epsilon: float = 1e-06, stash_type: int = 1> (add_412, "text.transformer.resblocks.3.ln_1.weight", "text.transformer.resblocks.3.ln_1.bias")
   [node_scaled_dot_product_attention_3_qkv_mm] node_scaled_dot_product_attention_3_qkv_mm_out = MatMul (layer_norm_6, val_13)
   [node_scaled_dot_product_attention_3_qkv_bias] node_scaled_dot_product_attention_3_qkv = Add (node_scaled_dot_product_attention_3_qkv_mm_out, "text.transformer.resblocks.3.attn.in_proj_bias")
   [node_scaled_dot_product_attention_3_qkv_split] node_scaled_dot_product_attention_3_q, node_scaled_dot_product_attention_3_k, node_scaled_dot_product_attention_3_v = Split <axis: int = -1> (node_scaled_dot_product_attention_3_qkv, attn3d_split_3x1152)
   scaled_dot_product_attention_3 = Attention <is_causal: int = 0, kv_num_heads: int = 16, q_num_heads: int = 16, qk_matmul_output_mode: int = 0, softcap: float = 0> (node_scaled_dot_product_attention_3_q, node_scaled_dot_product_attention_3_k, node_scaled_dot_product_attention_3_v)
   [node_scaled_dot_product_attention_3_out_mm] node_scaled_dot_product_attention_3_out_mm_out = MatMul (scaled_dot_product_attention_3, node_scaled_dot_product_attention_3_wo_t)
   [node_scaled_dot_product_attention_3_out_bias] node_scaled_dot_product_attention_3_out = Add (node_scaled_dot_product_attention_3_out_mm_out, "text.transformer.resblocks.3.attn.out_proj.bias")
   add_527 = Add (add_412, node_scaled_dot_product_attention_3_out)
   layer_norm_7 = LayerNormalization <axis: int = -1, epsilon: float = 1e-06, stash_type: int = 1> (add_527, "text.transformer.resblocks.3.ln_2.weight", "text.transformer.resblocks.3.ln_2.bias")
   val_92 = MatMul (layer_norm_7, val_14)
   linear_14 = Add (val_92, "text.transformer.resblocks.3.mlp.c_fc.bias")
   gelu_3 = Gelu <approximate: string = "tanh"> (linear_14)
   val_93 = MatMul (gelu_3, val_15)
   linear_15 = Add (val_93, "text.transformer.resblocks.3.mlp.c_proj.bias")
   add_548 = Add (add_527, linear_15)
   layer_norm_8 = LayerNormalization <axis: int = -1, epsilon: float = 1e-06, stash_type: int = 1> (add_548, "text.transformer.resblocks.4.ln_1.weight", "text.transformer.resblocks.4.ln_1.bias")
   [node_scaled_dot_product_attention_4_qkv_mm] node_scaled_dot_product_attention_4_qkv_mm_out = MatMul (layer_norm_8, val_16)
   [node_scaled_dot_product_attention_4_qkv_bias] node_scaled_dot_product_attention_4_qkv = Add (node_scaled_dot_product_attention_4_qkv_mm_out, "text.transformer.resblocks.4.attn.in_proj_bias")
   [node_scaled_dot_product_attention_4_qkv_split] node_scaled_dot_product_attention_4_q, node_scaled_dot_product_attention_4_k, node_scaled_dot_product_attention_4_v = Split <axis: int = -1> (node_scaled_dot_product_attention_4_qkv, attn3d_split_3x1152)
   scaled_dot_product_attention_4 = Attention <is_causal: int = 0, kv_num_heads: int = 16, q_num_heads: int = 16, qk_matmul_output_mode: int = 0, softcap: float = 0> (node_scaled_dot_product_attention_4_q, node_scaled_dot_product_attention_4_k, node_scaled_dot_product_attention_4_v)
   [node_scaled_dot_product_attention_4_out_mm] node_scaled_dot_product_attention_4_out_mm_out = MatMul (scaled_dot_product_attention_4, node_scaled_dot_product_attention_4_wo_t)
   [node_scaled_dot_product_attention_4_out_bias] node_scaled_dot_product_attention_4_out = Add (node_scaled_dot_product_attention_4_out_mm_out, "text.transformer.resblocks.4.attn.out_proj.bias")
   add_663 = Add (add_548, node_scaled_dot_product_attention_4_out)
   layer_norm_9 = LayerNormalization <axis: int = -1, epsilon: float = 1e-06, stash_type: int = 1> (add_663, "text.transformer.resblocks.4.ln_2.weight", "text.transformer.resblocks.4.ln_2.bias")
   val_94 = MatMul (layer_norm_9, val_17)
   linear_18 = Add (val_94, "text.transformer.resblocks.4.mlp.c_fc.bias")
   gelu_4 = Gelu <approximate: string = "tanh"> (linear_18)
   val_95 = MatMul (gelu_4, val_18)
   linear_19 = Add (val_95, "text.transformer.resblocks.4.mlp.c_proj.bias")
   add_684 = Add (add_663, linear_19)
   layer_norm_10 = LayerNormalization <axis: int = -1, epsilon: float = 1e-06, stash_type: int = 1> (add_684, "text.transformer.resblocks.5.ln_1.weight", "text.transformer.resblocks.5.ln_1.bias")
   [node_scaled_dot_product_attention_5_qkv_mm] node_scaled_dot_product_attention_5_qkv_mm_out = MatMul (layer_norm_10, val_19)
   [node_scaled_dot_product_attention_5_qkv_bias] node_scaled_dot_product_attention_5_qkv = Add (node_scaled_dot_product_attention_5_qkv_mm_out, "text.transformer.resblocks.5.attn.in_proj_bias")
   [node_scaled_dot_product_attention_5_qkv_split] node_scaled_dot_product_attention_5_q, node_scaled_dot_product_attention_5_k, node_scaled_dot_product_attention_5_v = Split <axis: int = -1> (node_scaled_dot_product_attention_5_qkv, attn3d_split_3x1152)
   scaled_dot_product_attention_5 = Attention <is_causal: int = 0, kv_num_heads: int = 16, q_num_heads: int = 16, qk_matmul_output_mode: int = 0, softcap: float = 0> (node_scaled_dot_product_attention_5_q, node_scaled_dot_product_attention_5_k, node_scaled_dot_product_attention_5_v)
   [node_scaled_dot_product_attention_5_out_mm] node_scaled_dot_product_attention_5_out_mm_out = MatMul (scaled_dot_product_attention_5, node_scaled_dot_product_attention_5_wo_t)
   [node_scaled_dot_product_attention_5_out_bias] node_scaled_dot_product_attention_5_out = Add (node_scaled_dot_product_attention_5_out_mm_out, "text.transformer.resblocks.5.attn.out_proj.bias")
   add_799 = Add (add_684, node_scaled_dot_product_attention_5_out)
   layer_norm_11 = LayerNormalization <axis: int = -1, epsilon: float = 1e-06, stash_type: int = 1> (add_799, "text.transformer.resblocks.5.ln_2.weight", "text.transformer.resblocks.5.ln_2.bias")
   val_96 = MatMul (layer_norm_11, val_20)
   linear_22 = Add (val_96, "text.transformer.resblocks.5.mlp.c_fc.bias")
   gelu_5 = Gelu <approximate: string = "tanh"> (linear_22)
   val_97 = MatMul (gelu_5, val_21)
   linear_23 = Add (val_97, "text.transformer.resblocks.5.mlp.c_proj.bias")
   add_820 = Add (add_799, linear_23)
   layer_norm_12 = LayerNormalization <axis: int = -1, epsilon: float = 1e-06, stash_type: int = 1> (add_820, "text.transformer.resblocks.6.ln_1.weight", "text.transformer.resblocks.6.ln_1.bias")
   [node_scaled_dot_product_attention_6_qkv_mm] node_scaled_dot_product_attention_6_qkv_mm_out = MatMul (layer_norm_12, val_22)
   [node_scaled_dot_product_attention_6_qkv_bias] node_scaled_dot_product_attention_6_qkv = Add (node_scaled_dot_product_attention_6_qkv_mm_out, "text.transformer.resblocks.6.attn.in_proj_bias")
   [node_scaled_dot_product_attention_6_qkv_split] node_scaled_dot_product_attention_6_q, node_scaled_dot_product_attention_6_k, node_scaled_dot_product_attention_6_v = Split <axis: int = -1> (node_scaled_dot_product_attention_6_qkv, attn3d_split_3x1152)
   scaled_dot_product_attention_6 = Attention <is_causal: int = 0, kv_num_heads: int = 16, q_num_heads: int = 16, qk_matmul_output_mode: int = 0, softcap: float = 0> (node_scaled_dot_product_attention_6_q, node_scaled_dot_product_attention_6_k, node_scaled_dot_product_attention_6_v)
   [node_scaled_dot_product_attention_6_out_mm] node_scaled_dot_product_attention_6_out_mm_out = MatMul (scaled_dot_product_attention_6, node_scaled_dot_product_attention_6_wo_t)
   [node_scaled_dot_product_attention_6_out_bias] node_scaled_dot_product_attention_6_out = Add (node_scaled_dot_product_attention_6_out_mm_out, "text.transformer.resblocks.6.attn.out_proj.bias")
   add_935 = Add (add_820, node_scaled_dot_product_attention_6_out)
   layer_norm_13 = LayerNormalization <axis: int = -1, epsilon: float = 1e-06, stash_type: int = 1> (add_935, "text.transformer.resblocks.6.ln_2.weight", "text.transformer.resblocks.6.ln_2.bias")
   val_98 = MatMul (layer_norm_13, val_23)
   linear_26 = Add (val_98, "text.transformer.resblocks.6.mlp.c_fc.bias")
   gelu_6 = Gelu <approximate: string = "tanh"> (linear_26)
   val_99 = MatMul (gelu_6, val_24)
   linear_27 = Add (val_99, "text.transformer.resblocks.6.mlp.c_proj.bias")
   add_956 = Add (add_935, linear_27)
   layer_norm_14 = LayerNormalization <axis: int = -1, epsilon: float = 1e-06, stash_type: int = 1> (add_956, "text.transformer.resblocks.7.ln_1.weight", "text.transformer.resblocks.7.ln_1.bias")
   [node_scaled_dot_product_attention_7_qkv_mm] node_scaled_dot_product_attention_7_qkv_mm_out = MatMul (layer_norm_14, val_25)
   [node_scaled_dot_product_attention_7_qkv_bias] node_scaled_dot_product_attention_7_qkv = Add (node_scaled_dot_product_attention_7_qkv_mm_out, "text.transformer.resblocks.7.attn.in_proj_bias")
   [node_scaled_dot_product_attention_7_qkv_split] node_scaled_dot_product_attention_7_q, node_scaled_dot_product_attention_7_k, node_scaled_dot_product_attention_7_v = Split <axis: int = -1> (node_scaled_dot_product_attention_7_qkv, attn3d_split_3x1152)
   scaled_dot_product_attention_7 = Attention <is_causal: int = 0, kv_num_heads: int = 16, q_num_heads: int = 16, qk_matmul_output_mode: int = 0, softcap: float = 0> (node_scaled_dot_product_attention_7_q, node_scaled_dot_product_attention_7_k, node_scaled_dot_product_attention_7_v)
   [node_scaled_dot_product_attention_7_out_mm] node_scaled_dot_product_attention_7_out_mm_out = MatMul (scaled_dot_product_attention_7, node_scaled_dot_product_attention_7_wo_t)
   [node_scaled_dot_product_attention_7_out_bias] node_scaled_dot_product_attention_7_out = Add (node_scaled_dot_product_attention_7_out_mm_out, "text.transformer.resblocks.7.attn.out_proj.bias")
   add_1071 = Add (add_956, node_scaled_dot_product_attention_7_out)
   layer_norm_15 = LayerNormalization <axis: int = -1, epsilon: float = 1e-06, stash_type: int = 1> (add_1071, "text.transformer.resblocks.7.ln_2.weight", "text.transformer.resblocks.7.ln_2.bias")
   val_100 = MatMul (layer_norm_15, val_26)
   linear_30 = Add (val_100, "text.transformer.resblocks.7.mlp.c_fc.bias")
   gelu_7 = Gelu <approximate: string = "tanh"> (linear_30)
   val_101 = MatMul (gelu_7, val_27)
   linear_31 = Add (val_101, "text.transformer.resblocks.7.mlp.c_proj.bias")
   add_1092 = Add (add_1071, linear_31)
   layer_norm_16 = LayerNormalization <axis: int = -1, epsilon: float = 1e-06, stash_type: int = 1> (add_1092, "text.transformer.resblocks.8.ln_1.weight", "text.transformer.resblocks.8.ln_1.bias")
   [node_scaled_dot_product_attention_8_qkv_mm] node_scaled_dot_product_attention_8_qkv_mm_out = MatMul (layer_norm_16, val_28)
   [node_scaled_dot_product_attention_8_qkv_bias] node_scaled_dot_product_attention_8_qkv = Add (node_scaled_dot_product_attention_8_qkv_mm_out, "text.transformer.resblocks.8.attn.in_proj_bias")
   [node_scaled_dot_product_attention_8_qkv_split] node_scaled_dot_product_attention_8_q, node_scaled_dot_product_attention_8_k, node_scaled_dot_product_attention_8_v = Split <axis: int = -1> (node_scaled_dot_product_attention_8_qkv, attn3d_split_3x1152)
   scaled_dot_product_attention_8 = Attention <is_causal: int = 0, kv_num_heads: int = 16, q_num_heads: int = 16, qk_matmul_output_mode: int = 0, softcap: float = 0> (node_scaled_dot_product_attention_8_q, node_scaled_dot_product_attention_8_k, node_scaled_dot_product_attention_8_v)
   [node_scaled_dot_product_attention_8_out_mm] node_scaled_dot_product_attention_8_out_mm_out = MatMul (scaled_dot_product_attention_8, node_scaled_dot_product_attention_8_wo_t)
   [node_scaled_dot_product_attention_8_out_bias] node_scaled_dot_product_attention_8_out = Add (node_scaled_dot_product_attention_8_out_mm_out, "text.transformer.resblocks.8.attn.out_proj.bias")
   add_1207 = Add (add_1092, node_scaled_dot_product_attention_8_out)
   layer_norm_17 = LayerNormalization <axis: int = -1, epsilon: float = 1e-06, stash_type: int = 1> (add_1207, "text.transformer.resblocks.8.ln_2.weight", "text.transformer.resblocks.8.ln_2.bias")
   val_102 = MatMul (layer_norm_17, val_29)
   linear_34 = Add (val_102, "text.transformer.resblocks.8.mlp.c_fc.bias")
   gelu_8 = Gelu <approximate: string = "tanh"> (linear_34)
   val_103 = MatMul (gelu_8, val_30)
   linear_35 = Add (val_103, "text.transformer.resblocks.8.mlp.c_proj.bias")
   add_1228 = Add (add_1207, linear_35)
   layer_norm_18 = LayerNormalization <axis: int = -1, epsilon: float = 1e-06, stash_type: int = 1> (add_1228, "text.transformer.resblocks.9.ln_1.weight", "text.transformer.resblocks.9.ln_1.bias")
   [node_scaled_dot_product_attention_9_qkv_mm] node_scaled_dot_product_attention_9_qkv_mm_out = MatMul (layer_norm_18, val_31)
   [node_scaled_dot_product_attention_9_qkv_bias] node_scaled_dot_product_attention_9_qkv = Add (node_scaled_dot_product_attention_9_qkv_mm_out, "text.transformer.resblocks.9.attn.in_proj_bias")
   [node_scaled_dot_product_attention_9_qkv_split] node_scaled_dot_product_attention_9_q, node_scaled_dot_product_attention_9_k, node_scaled_dot_product_attention_9_v = Split <axis: int = -1> (node_scaled_dot_product_attention_9_qkv, attn3d_split_3x1152)
   scaled_dot_product_attention_9 = Attention <is_causal: int = 0, kv_num_heads: int = 16, q_num_heads: int = 16, qk_matmul_output_mode: int = 0, softcap: float = 0> (node_scaled_dot_product_attention_9_q, node_scaled_dot_product_attention_9_k, node_scaled_dot_product_attention_9_v)
   [node_scaled_dot_product_attention_9_out_mm] node_scaled_dot_product_attention_9_out_mm_out = MatMul (scaled_dot_product_attention_9, node_scaled_dot_product_attention_9_wo_t)
   [node_scaled_dot_product_attention_9_out_bias] node_scaled_dot_product_attention_9_out = Add (node_scaled_dot_product_attention_9_out_mm_out, "text.transformer.resblocks.9.attn.out_proj.bias")
   add_1343 = Add (add_1228, node_scaled_dot_product_attention_9_out)
   layer_norm_19 = LayerNormalization <axis: int = -1, epsilon: float = 1e-06, stash_type: int = 1> (add_1343, "text.transformer.resblocks.9.ln_2.weight", "text.transformer.resblocks.9.ln_2.bias")
   val_104 = MatMul (layer_norm_19, val_32)
   linear_38 = Add (val_104, "text.transformer.resblocks.9.mlp.c_fc.bias")
   gelu_9 = Gelu <approximate: string = "tanh"> (linear_38)
   val_105 = MatMul (gelu_9, val_33)
   linear_39 = Add (val_105, "text.transformer.resblocks.9.mlp.c_proj.bias")
   add_1364 = Add (add_1343, linear_39)
   layer_norm_20 = LayerNormalization <axis: int = -1, epsilon: float = 1e-06, stash_type: int = 1> (add_1364, "text.transformer.resblocks.10.ln_1.weight", "text.transformer.resblocks.10.ln_1.bias")
   [node_scaled_dot_product_attention_10_qkv_mm] node_scaled_dot_product_attention_10_qkv_mm_out = MatMul (layer_norm_20, val_34)
   [node_scaled_dot_product_attention_10_qkv_bias] node_scaled_dot_product_attention_10_qkv = Add (node_scaled_dot_product_attention_10_qkv_mm_out, "text.transformer.resblocks.10.attn.in_proj_bias")
   [node_scaled_dot_product_attention_10_qkv_split] node_scaled_dot_product_attention_10_q, node_scaled_dot_product_attention_10_k, node_scaled_dot_product_attention_10_v = Split <axis: int = -1> (node_scaled_dot_product_attention_10_qkv, attn3d_split_3x1152)
   scaled_dot_product_attention_10 = Attention <is_causal: int = 0, kv_num_heads: int = 16, q_num_heads: int = 16, qk_matmul_output_mode: int = 0, softcap: float = 0> (node_scaled_dot_product_attention_10_q, node_scaled_dot_product_attention_10_k, node_scaled_dot_product_attention_10_v)
   [node_scaled_dot_product_attention_10_out_mm] node_scaled_dot_product_attention_10_out_mm_out = MatMul (scaled_dot_product_attention_10, node_scaled_dot_product_attention_10_wo_t)
   [node_scaled_dot_product_attention_10_out_bias] node_scaled_dot_product_attention_10_out = Add (node_scaled_dot_product_attention_10_out_mm_out, "text.transformer.resblocks.10.attn.out_proj.bias")
   add_1479 = Add (add_1364, node_scaled_dot_product_attention_10_out)
   layer_norm_21 = LayerNormalization <axis: int = -1, epsilon: float = 1e-06, stash_type: int = 1> (add_1479, "text.transformer.resblocks.10.ln_2.weight", "text.transformer.resblocks.10.ln_2.bias")
   val_106 = MatMul (layer_norm_21, val_35)
   linear_42 = Add (val_106, "text.transformer.resblocks.10.mlp.c_fc.bias")
   gelu_10 = Gelu <approximate: string = "tanh"> (linear_42)
   val_107 = MatMul (gelu_10, val_36)
   linear_43 = Add (val_107, "text.transformer.resblocks.10.mlp.c_proj.bias")
   add_1500 = Add (add_1479, linear_43)
   layer_norm_22 = LayerNormalization <axis: int = -1, epsilon: float = 1e-06, stash_type: int = 1> (add_1500, "text.transformer.resblocks.11.ln_1.weight", "text.transformer.resblocks.11.ln_1.bias")
   [node_scaled_dot_product_attention_11_qkv_mm] node_scaled_dot_product_attention_11_qkv_mm_out = MatMul (layer_norm_22, val_37)
   [node_scaled_dot_product_attention_11_qkv_bias] node_scaled_dot_product_attention_11_qkv = Add (node_scaled_dot_product_attention_11_qkv_mm_out, "text.transformer.resblocks.11.attn.in_proj_bias")
   [node_scaled_dot_product_attention_11_qkv_split] node_scaled_dot_product_attention_11_q, node_scaled_dot_product_attention_11_k, node_scaled_dot_product_attention_11_v = Split <axis: int = -1> (node_scaled_dot_product_attention_11_qkv, attn3d_split_3x1152)
   scaled_dot_product_attention_11 = Attention <is_causal: int = 0, kv_num_heads: int = 16, q_num_heads: int = 16, qk_matmul_output_mode: int = 0, softcap: float = 0> (node_scaled_dot_product_attention_11_q, node_scaled_dot_product_attention_11_k, node_scaled_dot_product_attention_11_v)
   [node_scaled_dot_product_attention_11_out_mm] node_scaled_dot_product_attention_11_out_mm_out = MatMul (scaled_dot_product_attention_11, node_scaled_dot_product_attention_11_wo_t)
   [node_scaled_dot_product_attention_11_out_bias] node_scaled_dot_product_attention_11_out = Add (node_scaled_dot_product_attention_11_out_mm_out, "text.transformer.resblocks.11.attn.out_proj.bias")
   add_1615 = Add (add_1500, node_scaled_dot_product_attention_11_out)
   layer_norm_23 = LayerNormalization <axis: int = -1, epsilon: float = 1e-06, stash_type: int = 1> (add_1615, "text.transformer.resblocks.11.ln_2.weight", "text.transformer.resblocks.11.ln_2.bias")
   val_108 = MatMul (layer_norm_23, val_38)
   linear_46 = Add (val_108, "text.transformer.resblocks.11.mlp.c_fc.bias")
   gelu_11 = Gelu <approximate: string = "tanh"> (linear_46)
   val_109 = MatMul (gelu_11, val_39)
   linear_47 = Add (val_109, "text.transformer.resblocks.11.mlp.c_proj.bias")
   add_1636 = Add (add_1615, linear_47)
   layer_norm_24 = LayerNormalization <axis: int = -1, epsilon: float = 1e-06, stash_type: int = 1> (add_1636, "text.transformer.resblocks.12.ln_1.weight", "text.transformer.resblocks.12.ln_1.bias")
   [node_scaled_dot_product_attention_12_qkv_mm] node_scaled_dot_product_attention_12_qkv_mm_out = MatMul (layer_norm_24, val_40)
   [node_scaled_dot_product_attention_12_qkv_bias] node_scaled_dot_product_attention_12_qkv = Add (node_scaled_dot_product_attention_12_qkv_mm_out, "text.transformer.resblocks.12.attn.in_proj_bias")
   [node_scaled_dot_product_attention_12_qkv_split] node_scaled_dot_product_attention_12_q, node_scaled_dot_product_attention_12_k, node_scaled_dot_product_attention_12_v = Split <axis: int = -1> (node_scaled_dot_product_attention_12_qkv, attn3d_split_3x1152)
   scaled_dot_product_attention_12 = Attention <is_causal: int = 0, kv_num_heads: int = 16, q_num_heads: int = 16, qk_matmul_output_mode: int = 0, softcap: float = 0> (node_scaled_dot_product_attention_12_q, node_scaled_dot_product_attention_12_k, node_scaled_dot_product_attention_12_v)
   [node_scaled_dot_product_attention_12_out_mm] node_scaled_dot_product_attention_12_out_mm_out = MatMul (scaled_dot_product_attention_12, node_scaled_dot_product_attention_12_wo_t)
   [node_scaled_dot_product_attention_12_out_bias] node_scaled_dot_product_attention_12_out = Add (node_scaled_dot_product_attention_12_out_mm_out, "text.transformer.resblocks.12.attn.out_proj.bias")
   add_1751 = Add (add_1636, node_scaled_dot_product_attention_12_out)
   layer_norm_25 = LayerNormalization <axis: int = -1, epsilon: float = 1e-06, stash_type: int = 1> (add_1751, "text.transformer.resblocks.12.ln_2.weight", "text.transformer.resblocks.12.ln_2.bias")
   val_110 = MatMul (layer_norm_25, val_41)
   linear_50 = Add (val_110, "text.transformer.resblocks.12.mlp.c_fc.bias")
   gelu_12 = Gelu <approximate: string = "tanh"> (linear_50)
   val_111 = MatMul (gelu_12, val_42)
   linear_51 = Add (val_111, "text.transformer.resblocks.12.mlp.c_proj.bias")
   add_1772 = Add (add_1751, linear_51)
   layer_norm_26 = LayerNormalization <axis: int = -1, epsilon: float = 1e-06, stash_type: int = 1> (add_1772, "text.transformer.resblocks.13.ln_1.weight", "text.transformer.resblocks.13.ln_1.bias")
   [node_scaled_dot_product_attention_13_qkv_mm] node_scaled_dot_product_attention_13_qkv_mm_out = MatMul (layer_norm_26, val_43)
   [node_scaled_dot_product_attention_13_qkv_bias] node_scaled_dot_product_attention_13_qkv = Add (node_scaled_dot_product_attention_13_qkv_mm_out, "text.transformer.resblocks.13.attn.in_proj_bias")
   [node_scaled_dot_product_attention_13_qkv_split] node_scaled_dot_product_attention_13_q, node_scaled_dot_product_attention_13_k, node_scaled_dot_product_attention_13_v = Split <axis: int = -1> (node_scaled_dot_product_attention_13_qkv, attn3d_split_3x1152)
   scaled_dot_product_attention_13 = Attention <is_causal: int = 0, kv_num_heads: int = 16, q_num_heads: int = 16, qk_matmul_output_mode: int = 0, softcap: float = 0> (node_scaled_dot_product_attention_13_q, node_scaled_dot_product_attention_13_k, node_scaled_dot_product_attention_13_v)
   [node_scaled_dot_product_attention_13_out_mm] node_scaled_dot_product_attention_13_out_mm_out = MatMul (scaled_dot_product_attention_13, node_scaled_dot_product_attention_13_wo_t)
   [node_scaled_dot_product_attention_13_out_bias] node_scaled_dot_product_attention_13_out = Add (node_scaled_dot_product_attention_13_out_mm_out, "text.transformer.resblocks.13.attn.out_proj.bias")
   add_1887 = Add (add_1772, node_scaled_dot_product_attention_13_out)
   layer_norm_27 = LayerNormalization <axis: int = -1, epsilon: float = 1e-06, stash_type: int = 1> (add_1887, "text.transformer.resblocks.13.ln_2.weight", "text.transformer.resblocks.13.ln_2.bias")
   val_112 = MatMul (layer_norm_27, val_44)
   linear_54 = Add (val_112, "text.transformer.resblocks.13.mlp.c_fc.bias")
   gelu_13 = Gelu <approximate: string = "tanh"> (linear_54)
   val_113 = MatMul (gelu_13, val_45)
   linear_55 = Add (val_113, "text.transformer.resblocks.13.mlp.c_proj.bias")
   add_1908 = Add (add_1887, linear_55)
   layer_norm_28 = LayerNormalization <axis: int = -1, epsilon: float = 1e-06, stash_type: int = 1> (add_1908, "text.transformer.resblocks.14.ln_1.weight", "text.transformer.resblocks.14.ln_1.bias")
   [node_scaled_dot_product_attention_14_qkv_mm] node_scaled_dot_product_attention_14_qkv_mm_out = MatMul (layer_norm_28, val_46)
   [node_scaled_dot_product_attention_14_qkv_bias] node_scaled_dot_product_attention_14_qkv = Add (node_scaled_dot_product_attention_14_qkv_mm_out, "text.transformer.resblocks.14.attn.in_proj_bias")
   [node_scaled_dot_product_attention_14_qkv_split] node_scaled_dot_product_attention_14_q, node_scaled_dot_product_attention_14_k, node_scaled_dot_product_attention_14_v = Split <axis: int = -1> (node_scaled_dot_product_attention_14_qkv, attn3d_split_3x1152)
   scaled_dot_product_attention_14 = Attention <is_causal: int = 0, kv_num_heads: int = 16, q_num_heads: int = 16, qk_matmul_output_mode: int = 0, softcap: float = 0> (node_scaled_dot_product_attention_14_q, node_scaled_dot_product_attention_14_k, node_scaled_dot_product_attention_14_v)
   [node_scaled_dot_product_attention_14_out_mm] node_scaled_dot_product_attention_14_out_mm_out = MatMul (scaled_dot_product_attention_14, node_scaled_dot_product_attention_14_wo_t)
   [node_scaled_dot_product_attention_14_out_bias] node_scaled_dot_product_attention_14_out = Add (node_scaled_dot_product_attention_14_out_mm_out, "text.transformer.resblocks.14.attn.out_proj.bias")
   add_2023 = Add (add_1908, node_scaled_dot_product_attention_14_out)
   layer_norm_29 = LayerNormalization <axis: int = -1, epsilon: float = 1e-06, stash_type: int = 1> (add_2023, "text.transformer.resblocks.14.ln_2.weight", "text.transformer.resblocks.14.ln_2.bias")
   val_114 = MatMul (layer_norm_29, val_47)
   linear_58 = Add (val_114, "text.transformer.resblocks.14.mlp.c_fc.bias")
   gelu_14 = Gelu <approximate: string = "tanh"> (linear_58)
   val_115 = MatMul (gelu_14, val_48)
   linear_59 = Add (val_115, "text.transformer.resblocks.14.mlp.c_proj.bias")
   add_2044 = Add (add_2023, linear_59)
   layer_norm_30 = LayerNormalization <axis: int = -1, epsilon: float = 1e-06, stash_type: int = 1> (add_2044, "text.transformer.resblocks.15.ln_1.weight", "text.transformer.resblocks.15.ln_1.bias")
   [node_scaled_dot_product_attention_15_qkv_mm] node_scaled_dot_product_attention_15_qkv_mm_out = MatMul (layer_norm_30, val_49)
   [node_scaled_dot_product_attention_15_qkv_bias] node_scaled_dot_product_attention_15_qkv = Add (node_scaled_dot_product_attention_15_qkv_mm_out, "text.transformer.resblocks.15.attn.in_proj_bias")
   [node_scaled_dot_product_attention_15_qkv_split] node_scaled_dot_product_attention_15_q, node_scaled_dot_product_attention_15_k, node_scaled_dot_product_attention_15_v = Split <axis: int = -1> (node_scaled_dot_product_attention_15_qkv, attn3d_split_3x1152)
   scaled_dot_product_attention_15 = Attention <is_causal: int = 0, kv_num_heads: int = 16, q_num_heads: int = 16, qk_matmul_output_mode: int = 0, softcap: float = 0> (node_scaled_dot_product_attention_15_q, node_scaled_dot_product_attention_15_k, node_scaled_dot_product_attention_15_v)
   [node_scaled_dot_product_attention_15_out_mm] node_scaled_dot_product_attention_15_out_mm_out = MatMul (scaled_dot_product_attention_15, node_scaled_dot_product_attention_15_wo_t)
   [node_scaled_dot_product_attention_15_out_bias] node_scaled_dot_product_attention_15_out = Add (node_scaled_dot_product_attention_15_out_mm_out, "text.transformer.resblocks.15.attn.out_proj.bias")
   add_2159 = Add (add_2044, node_scaled_dot_product_attention_15_out)
   layer_norm_31 = LayerNormalization <axis: int = -1, epsilon: float = 1e-06, stash_type: int = 1> (add_2159, "text.transformer.resblocks.15.ln_2.weight", "text.transformer.resblocks.15.ln_2.bias")
   val_116 = MatMul (layer_norm_31, val_50)
   linear_62 = Add (val_116, "text.transformer.resblocks.15.mlp.c_fc.bias")
   gelu_15 = Gelu <approximate: string = "tanh"> (linear_62)
   val_117 = MatMul (gelu_15, val_51)
   linear_63 = Add (val_117, "text.transformer.resblocks.15.mlp.c_proj.bias")
   add_2180 = Add (add_2159, linear_63)
   layer_norm_32 = LayerNormalization <axis: int = -1, epsilon: float = 1e-06, stash_type: int = 1> (add_2180, "text.transformer.resblocks.16.ln_1.weight", "text.transformer.resblocks.16.ln_1.bias")
   [node_scaled_dot_product_attention_16_qkv_mm] node_scaled_dot_product_attention_16_qkv_mm_out = MatMul (layer_norm_32, val_52)
   [node_scaled_dot_product_attention_16_qkv_bias] node_scaled_dot_product_attention_16_qkv = Add (node_scaled_dot_product_attention_16_qkv_mm_out, "text.transformer.resblocks.16.attn.in_proj_bias")
   [node_scaled_dot_product_attention_16_qkv_split] node_scaled_dot_product_attention_16_q, node_scaled_dot_product_attention_16_k, node_scaled_dot_product_attention_16_v = Split <axis: int = -1> (node_scaled_dot_product_attention_16_qkv, attn3d_split_3x1152)
   scaled_dot_product_attention_16 = Attention <is_causal: int = 0, kv_num_heads: int = 16, q_num_heads: int = 16, qk_matmul_output_mode: int = 0, softcap: float = 0> (node_scaled_dot_product_attention_16_q, node_scaled_dot_product_attention_16_k, node_scaled_dot_product_attention_16_v)
   [node_scaled_dot_product_attention_16_out_mm] node_scaled_dot_product_attention_16_out_mm_out = MatMul (scaled_dot_product_attention_16, node_scaled_dot_product_attention_16_wo_t)
   [node_scaled_dot_product_attention_16_out_bias] node_scaled_dot_product_attention_16_out = Add (node_scaled_dot_product_attention_16_out_mm_out, "text.transformer.resblocks.16.attn.out_proj.bias")
   add_2295 = Add (add_2180, node_scaled_dot_product_attention_16_out)
   layer_norm_33 = LayerNormalization <axis: int = -1, epsilon: float = 1e-06, stash_type: int = 1> (add_2295, "text.transformer.resblocks.16.ln_2.weight", "text.transformer.resblocks.16.ln_2.bias")
   val_118 = MatMul (layer_norm_33, val_53)
   linear_66 = Add (val_118, "text.transformer.resblocks.16.mlp.c_fc.bias")
   gelu_16 = Gelu <approximate: string = "tanh"> (linear_66)
   val_119 = MatMul (gelu_16, val_54)
   linear_67 = Add (val_119, "text.transformer.resblocks.16.mlp.c_proj.bias")
   add_2316 = Add (add_2295, linear_67)
   layer_norm_34 = LayerNormalization <axis: int = -1, epsilon: float = 1e-06, stash_type: int = 1> (add_2316, "text.transformer.resblocks.17.ln_1.weight", "text.transformer.resblocks.17.ln_1.bias")
   [node_scaled_dot_product_attention_17_qkv_mm] node_scaled_dot_product_attention_17_qkv_mm_out = MatMul (layer_norm_34, val_55)
   [node_scaled_dot_product_attention_17_qkv_bias] node_scaled_dot_product_attention_17_qkv = Add (node_scaled_dot_product_attention_17_qkv_mm_out, "text.transformer.resblocks.17.attn.in_proj_bias")
   [node_scaled_dot_product_attention_17_qkv_split] node_scaled_dot_product_attention_17_q, node_scaled_dot_product_attention_17_k, node_scaled_dot_product_attention_17_v = Split <axis: int = -1> (node_scaled_dot_product_attention_17_qkv, attn3d_split_3x1152)
   scaled_dot_product_attention_17 = Attention <is_causal: int = 0, kv_num_heads: int = 16, q_num_heads: int = 16, qk_matmul_output_mode: int = 0, softcap: float = 0> (node_scaled_dot_product_attention_17_q, node_scaled_dot_product_attention_17_k, node_scaled_dot_product_attention_17_v)
   [node_scaled_dot_product_attention_17_out_mm] node_scaled_dot_product_attention_17_out_mm_out = MatMul (scaled_dot_product_attention_17, node_scaled_dot_product_attention_17_wo_t)
   [node_scaled_dot_product_attention_17_out_bias] node_scaled_dot_product_attention_17_out = Add (node_scaled_dot_product_attention_17_out_mm_out, "text.transformer.resblocks.17.attn.out_proj.bias")
   add_2431 = Add (add_2316, node_scaled_dot_product_attention_17_out)
   layer_norm_35 = LayerNormalization <axis: int = -1, epsilon: float = 1e-06, stash_type: int = 1> (add_2431, "text.transformer.resblocks.17.ln_2.weight", "text.transformer.resblocks.17.ln_2.bias")
   val_120 = MatMul (layer_norm_35, val_56)
   linear_70 = Add (val_120, "text.transformer.resblocks.17.mlp.c_fc.bias")
   gelu_17 = Gelu <approximate: string = "tanh"> (linear_70)
   val_121 = MatMul (gelu_17, val_57)
   linear_71 = Add (val_121, "text.transformer.resblocks.17.mlp.c_proj.bias")
   add_2452 = Add (add_2431, linear_71)
   layer_norm_36 = LayerNormalization <axis: int = -1, epsilon: float = 1e-06, stash_type: int = 1> (add_2452, "text.transformer.resblocks.18.ln_1.weight", "text.transformer.resblocks.18.ln_1.bias")
   [node_scaled_dot_product_attention_18_qkv_mm] node_scaled_dot_product_attention_18_qkv_mm_out = MatMul (layer_norm_36, val_58)
   [node_scaled_dot_product_attention_18_qkv_bias] node_scaled_dot_product_attention_18_qkv = Add (node_scaled_dot_product_attention_18_qkv_mm_out, "text.transformer.resblocks.18.attn.in_proj_bias")
   [node_scaled_dot_product_attention_18_qkv_split] node_scaled_dot_product_attention_18_q, node_scaled_dot_product_attention_18_k, node_scaled_dot_product_attention_18_v = Split <axis: int = -1> (node_scaled_dot_product_attention_18_qkv, attn3d_split_3x1152)
   scaled_dot_product_attention_18 = Attention <is_causal: int = 0, kv_num_heads: int = 16, q_num_heads: int = 16, qk_matmul_output_mode: int = 0, softcap: float = 0> (node_scaled_dot_product_attention_18_q, node_scaled_dot_product_attention_18_k, node_scaled_dot_product_attention_18_v)
   [node_scaled_dot_product_attention_18_out_mm] node_scaled_dot_product_attention_18_out_mm_out = MatMul (scaled_dot_product_attention_18, node_scaled_dot_product_attention_18_wo_t)
   [node_scaled_dot_product_attention_18_out_bias] node_scaled_dot_product_attention_18_out = Add (node_scaled_dot_product_attention_18_out_mm_out, "text.transformer.resblocks.18.attn.out_proj.bias")
   add_2567 = Add (add_2452, node_scaled_dot_product_attention_18_out)
   layer_norm_37 = LayerNormalization <axis: int = -1, epsilon: float = 1e-06, stash_type: int = 1> (add_2567, "text.transformer.resblocks.18.ln_2.weight", "text.transformer.resblocks.18.ln_2.bias")
   val_122 = MatMul (layer_norm_37, val_59)
   linear_74 = Add (val_122, "text.transformer.resblocks.18.mlp.c_fc.bias")
   gelu_18 = Gelu <approximate: string = "tanh"> (linear_74)
   val_123 = MatMul (gelu_18, val_60)
   linear_75 = Add (val_123, "text.transformer.resblocks.18.mlp.c_proj.bias")
   add_2588 = Add (add_2567, linear_75)
   layer_norm_38 = LayerNormalization <axis: int = -1, epsilon: float = 1e-06, stash_type: int = 1> (add_2588, "text.transformer.resblocks.19.ln_1.weight", "text.transformer.resblocks.19.ln_1.bias")
   [node_scaled_dot_product_attention_19_qkv_mm] node_scaled_dot_product_attention_19_qkv_mm_out = MatMul (layer_norm_38, val_61)
   [node_scaled_dot_product_attention_19_qkv_bias] node_scaled_dot_product_attention_19_qkv = Add (node_scaled_dot_product_attention_19_qkv_mm_out, "text.transformer.resblocks.19.attn.in_proj_bias")
   [node_scaled_dot_product_attention_19_qkv_split] node_scaled_dot_product_attention_19_q, node_scaled_dot_product_attention_19_k, node_scaled_dot_product_attention_19_v = Split <axis: int = -1> (node_scaled_dot_product_attention_19_qkv, attn3d_split_3x1152)
   scaled_dot_product_attention_19 = Attention <is_causal: int = 0, kv_num_heads: int = 16, q_num_heads: int = 16, qk_matmul_output_mode: int = 0, softcap: float = 0> (node_scaled_dot_product_attention_19_q, node_scaled_dot_product_attention_19_k, node_scaled_dot_product_attention_19_v)
   [node_scaled_dot_product_attention_19_out_mm] node_scaled_dot_product_attention_19_out_mm_out = MatMul (scaled_dot_product_attention_19, node_scaled_dot_product_attention_19_wo_t)
   [node_scaled_dot_product_attention_19_out_bias] node_scaled_dot_product_attention_19_out = Add (node_scaled_dot_product_attention_19_out_mm_out, "text.transformer.resblocks.19.attn.out_proj.bias")
   add_2703 = Add (add_2588, node_scaled_dot_product_attention_19_out)
   layer_norm_39 = LayerNormalization <axis: int = -1, epsilon: float = 1e-06, stash_type: int = 1> (add_2703, "text.transformer.resblocks.19.ln_2.weight", "text.transformer.resblocks.19.ln_2.bias")
   val_124 = MatMul (layer_norm_39, val_62)
   linear_78 = Add (val_124, "text.transformer.resblocks.19.mlp.c_fc.bias")
   gelu_19 = Gelu <approximate: string = "tanh"> (linear_78)
   val_125 = MatMul (gelu_19, val_63)
   linear_79 = Add (val_125, "text.transformer.resblocks.19.mlp.c_proj.bias")
   add_2724 = Add (add_2703, linear_79)
   layer_norm_40 = LayerNormalization <axis: int = -1, epsilon: float = 1e-06, stash_type: int = 1> (add_2724, "text.transformer.resblocks.20.ln_1.weight", "text.transformer.resblocks.20.ln_1.bias")
   [node_scaled_dot_product_attention_20_qkv_mm] node_scaled_dot_product_attention_20_qkv_mm_out = MatMul (layer_norm_40, val_64)
   [node_scaled_dot_product_attention_20_qkv_bias] node_scaled_dot_product_attention_20_qkv = Add (node_scaled_dot_product_attention_20_qkv_mm_out, "text.transformer.resblocks.20.attn.in_proj_bias")
   [node_scaled_dot_product_attention_20_qkv_split] node_scaled_dot_product_attention_20_q, node_scaled_dot_product_attention_20_k, node_scaled_dot_product_attention_20_v = Split <axis: int = -1> (node_scaled_dot_product_attention_20_qkv, attn3d_split_3x1152)
   scaled_dot_product_attention_20 = Attention <is_causal: int = 0, kv_num_heads: int = 16, q_num_heads: int = 16, qk_matmul_output_mode: int = 0, softcap: float = 0> (node_scaled_dot_product_attention_20_q, node_scaled_dot_product_attention_20_k, node_scaled_dot_product_attention_20_v)
   [node_scaled_dot_product_attention_20_out_mm] node_scaled_dot_product_attention_20_out_mm_out = MatMul (scaled_dot_product_attention_20, node_scaled_dot_product_attention_20_wo_t)
   [node_scaled_dot_product_attention_20_out_bias] node_scaled_dot_product_attention_20_out = Add (node_scaled_dot_product_attention_20_out_mm_out, "text.transformer.resblocks.20.attn.out_proj.bias")
   add_2839 = Add (add_2724, node_scaled_dot_product_attention_20_out)
   layer_norm_41 = LayerNormalization <axis: int = -1, epsilon: float = 1e-06, stash_type: int = 1> (add_2839, "text.transformer.resblocks.20.ln_2.weight", "text.transformer.resblocks.20.ln_2.bias")
   val_126 = MatMul (layer_norm_41, val_65)
   linear_82 = Add (val_126, "text.transformer.resblocks.20.mlp.c_fc.bias")
   gelu_20 = Gelu <approximate: string = "tanh"> (linear_82)
   val_127 = MatMul (gelu_20, val_66)
   linear_83 = Add (val_127, "text.transformer.resblocks.20.mlp.c_proj.bias")
   add_2860 = Add (add_2839, linear_83)
   layer_norm_42 = LayerNormalization <axis: int = -1, epsilon: float = 1e-06, stash_type: int = 1> (add_2860, "text.transformer.resblocks.21.ln_1.weight", "text.transformer.resblocks.21.ln_1.bias")
   [node_scaled_dot_product_attention_21_qkv_mm] node_scaled_dot_product_attention_21_qkv_mm_out = MatMul (layer_norm_42, val_67)
   [node_scaled_dot_product_attention_21_qkv_bias] node_scaled_dot_product_attention_21_qkv = Add (node_scaled_dot_product_attention_21_qkv_mm_out, "text.transformer.resblocks.21.attn.in_proj_bias")
   [node_scaled_dot_product_attention_21_qkv_split] node_scaled_dot_product_attention_21_q, node_scaled_dot_product_attention_21_k, node_scaled_dot_product_attention_21_v = Split <axis: int = -1> (node_scaled_dot_product_attention_21_qkv, attn3d_split_3x1152)
   scaled_dot_product_attention_21 = Attention <is_causal: int = 0, kv_num_heads: int = 16, q_num_heads: int = 16, qk_matmul_output_mode: int = 0, softcap: float = 0> (node_scaled_dot_product_attention_21_q, node_scaled_dot_product_attention_21_k, node_scaled_dot_product_attention_21_v)
   [node_scaled_dot_product_attention_21_out_mm] node_scaled_dot_product_attention_21_out_mm_out = MatMul (scaled_dot_product_attention_21, node_scaled_dot_product_attention_21_wo_t)
   [node_scaled_dot_product_attention_21_out_bias] node_scaled_dot_product_attention_21_out = Add (node_scaled_dot_product_attention_21_out_mm_out, "text.transformer.resblocks.21.attn.out_proj.bias")
   add_2975 = Add (add_2860, node_scaled_dot_product_attention_21_out)
   layer_norm_43 = LayerNormalization <axis: int = -1, epsilon: float = 1e-06, stash_type: int = 1> (add_2975, "text.transformer.resblocks.21.ln_2.weight", "text.transformer.resblocks.21.ln_2.bias")
   val_128 = MatMul (layer_norm_43, val_68)
   linear_86 = Add (val_128, "text.transformer.resblocks.21.mlp.c_fc.bias")
   gelu_21 = Gelu <approximate: string = "tanh"> (linear_86)
   val_129 = MatMul (gelu_21, val_69)
   linear_87 = Add (val_129, "text.transformer.resblocks.21.mlp.c_proj.bias")
   add_2996 = Add (add_2975, linear_87)
   layer_norm_44 = LayerNormalization <axis: int = -1, epsilon: float = 1e-06, stash_type: int = 1> (add_2996, "text.transformer.resblocks.22.ln_1.weight", "text.transformer.resblocks.22.ln_1.bias")
   [node_scaled_dot_product_attention_22_qkv_mm] node_scaled_dot_product_attention_22_qkv_mm_out = MatMul (layer_norm_44, val_70)
   [node_scaled_dot_product_attention_22_qkv_bias] node_scaled_dot_product_attention_22_qkv = Add (node_scaled_dot_product_attention_22_qkv_mm_out, "text.transformer.resblocks.22.attn.in_proj_bias")
   [node_scaled_dot_product_attention_22_qkv_split] node_scaled_dot_product_attention_22_q, node_scaled_dot_product_attention_22_k, node_scaled_dot_product_attention_22_v = Split <axis: int = -1> (node_scaled_dot_product_attention_22_qkv, attn3d_split_3x1152)
   scaled_dot_product_attention_22 = Attention <is_causal: int = 0, kv_num_heads: int = 16, q_num_heads: int = 16, qk_matmul_output_mode: int = 0, softcap: float = 0> (node_scaled_dot_product_attention_22_q, node_scaled_dot_product_attention_22_k, node_scaled_dot_product_attention_22_v)
   [node_scaled_dot_product_attention_22_out_mm] node_scaled_dot_product_attention_22_out_mm_out = MatMul (scaled_dot_product_attention_22, node_scaled_dot_product_attention_22_wo_t)
   [node_scaled_dot_product_attention_22_out_bias] node_scaled_dot_product_attention_22_out = Add (node_scaled_dot_product_attention_22_out_mm_out, "text.transformer.resblocks.22.attn.out_proj.bias")
   add_3111 = Add (add_2996, node_scaled_dot_product_attention_22_out)
   layer_norm_45 = LayerNormalization <axis: int = -1, epsilon: float = 1e-06, stash_type: int = 1> (add_3111, "text.transformer.resblocks.22.ln_2.weight", "text.transformer.resblocks.22.ln_2.bias")
   val_130 = MatMul (layer_norm_45, val_71)
   linear_90 = Add (val_130, "text.transformer.resblocks.22.mlp.c_fc.bias")
   gelu_22 = Gelu <approximate: string = "tanh"> (linear_90)
   val_131 = MatMul (gelu_22, val_72)
   linear_91 = Add (val_131, "text.transformer.resblocks.22.mlp.c_proj.bias")
   add_3132 = Add (add_3111, linear_91)
   layer_norm_46 = LayerNormalization <axis: int = -1, epsilon: float = 1e-06, stash_type: int = 1> (add_3132, "text.transformer.resblocks.23.ln_1.weight", "text.transformer.resblocks.23.ln_1.bias")
   [node_scaled_dot_product_attention_23_qkv_mm] node_scaled_dot_product_attention_23_qkv_mm_out = MatMul (layer_norm_46, val_73)
   [node_scaled_dot_product_attention_23_qkv_bias] node_scaled_dot_product_attention_23_qkv = Add (node_scaled_dot_product_attention_23_qkv_mm_out, "text.transformer.resblocks.23.attn.in_proj_bias")
   [node_scaled_dot_product_attention_23_qkv_split] node_scaled_dot_product_attention_23_q, node_scaled_dot_product_attention_23_k, node_scaled_dot_product_attention_23_v = Split <axis: int = -1> (node_scaled_dot_product_attention_23_qkv, attn3d_split_3x1152)
   scaled_dot_product_attention_23 = Attention <is_causal: int = 0, kv_num_heads: int = 16, q_num_heads: int = 16, qk_matmul_output_mode: int = 0, softcap: float = 0> (node_scaled_dot_product_attention_23_q, node_scaled_dot_product_attention_23_k, node_scaled_dot_product_attention_23_v)
   [node_scaled_dot_product_attention_23_out_mm] node_scaled_dot_product_attention_23_out_mm_out = MatMul (scaled_dot_product_attention_23, node_scaled_dot_product_attention_23_wo_t)
   [node_scaled_dot_product_attention_23_out_bias] node_scaled_dot_product_attention_23_out = Add (node_scaled_dot_product_attention_23_out_mm_out, "text.transformer.resblocks.23.attn.out_proj.bias")
   add_3247 = Add (add_3132, node_scaled_dot_product_attention_23_out)
   layer_norm_47 = LayerNormalization <axis: int = -1, epsilon: float = 1e-06, stash_type: int = 1> (add_3247, "text.transformer.resblocks.23.ln_2.weight", "text.transformer.resblocks.23.ln_2.bias")
   val_132 = MatMul (layer_norm_47, val_74)
   linear_94 = Add (val_132, "text.transformer.resblocks.23.mlp.c_fc.bias")
   gelu_23 = Gelu <approximate: string = "tanh"> (linear_94)
   val_133 = MatMul (gelu_23, val_75)
   linear_95 = Add (val_133, "text.transformer.resblocks.23.mlp.c_proj.bias")
   add_3268 = Add (add_3247, linear_95)
   layer_norm_48 = LayerNormalization <axis: int = -1, epsilon: float = 1e-06, stash_type: int = 1> (add_3268, "text.transformer.resblocks.24.ln_1.weight", "text.transformer.resblocks.24.ln_1.bias")
   [node_scaled_dot_product_attention_24_qkv_mm] node_scaled_dot_product_attention_24_qkv_mm_out = MatMul (layer_norm_48, val_76)
   [node_scaled_dot_product_attention_24_qkv_bias] node_scaled_dot_product_attention_24_qkv = Add (node_scaled_dot_product_attention_24_qkv_mm_out, "text.transformer.resblocks.24.attn.in_proj_bias")
   [node_scaled_dot_product_attention_24_qkv_split] node_scaled_dot_product_attention_24_q, node_scaled_dot_product_attention_24_k, node_scaled_dot_product_attention_24_v = Split <axis: int = -1> (node_scaled_dot_product_attention_24_qkv, attn3d_split_3x1152)
   scaled_dot_product_attention_24 = Attention <is_causal: int = 0, kv_num_heads: int = 16, q_num_heads: int = 16, qk_matmul_output_mode: int = 0, softcap: float = 0> (node_scaled_dot_product_attention_24_q, node_scaled_dot_product_attention_24_k, node_scaled_dot_product_attention_24_v)
   [node_scaled_dot_product_attention_24_out_mm] node_scaled_dot_product_attention_24_out_mm_out = MatMul (scaled_dot_product_attention_24, node_scaled_dot_product_attention_24_wo_t)
   [node_scaled_dot_product_attention_24_out_bias] node_scaled_dot_product_attention_24_out = Add (node_scaled_dot_product_attention_24_out_mm_out, "text.transformer.resblocks.24.attn.out_proj.bias")
   add_3383 = Add (add_3268, node_scaled_dot_product_attention_24_out)
   layer_norm_49 = LayerNormalization <axis: int = -1, epsilon: float = 1e-06, stash_type: int = 1> (add_3383, "text.transformer.resblocks.24.ln_2.weight", "text.transformer.resblocks.24.ln_2.bias")
   val_134 = MatMul (layer_norm_49, val_77)
   linear_98 = Add (val_134, "text.transformer.resblocks.24.mlp.c_fc.bias")
   gelu_24 = Gelu <approximate: string = "tanh"> (linear_98)
   val_135 = MatMul (gelu_24, val_78)
   linear_99 = Add (val_135, "text.transformer.resblocks.24.mlp.c_proj.bias")
   add_3404 = Add (add_3383, linear_99)
   layer_norm_50 = LayerNormalization <axis: int = -1, epsilon: float = 1e-06, stash_type: int = 1> (add_3404, "text.transformer.resblocks.25.ln_1.weight", "text.transformer.resblocks.25.ln_1.bias")
   [node_scaled_dot_product_attention_25_qkv_mm] node_scaled_dot_product_attention_25_qkv_mm_out = MatMul (layer_norm_50, val_79)
   [node_scaled_dot_product_attention_25_qkv_bias] node_scaled_dot_product_attention_25_qkv = Add (node_scaled_dot_product_attention_25_qkv_mm_out, "text.transformer.resblocks.25.attn.in_proj_bias")
   [node_scaled_dot_product_attention_25_qkv_split] node_scaled_dot_product_attention_25_q, node_scaled_dot_product_attention_25_k, node_scaled_dot_product_attention_25_v = Split <axis: int = -1> (node_scaled_dot_product_attention_25_qkv, attn3d_split_3x1152)
   scaled_dot_product_attention_25 = Attention <is_causal: int = 0, kv_num_heads: int = 16, q_num_heads: int = 16, qk_matmul_output_mode: int = 0, softcap: float = 0> (node_scaled_dot_product_attention_25_q, node_scaled_dot_product_attention_25_k, node_scaled_dot_product_attention_25_v)
   [node_scaled_dot_product_attention_25_out_mm] node_scaled_dot_product_attention_25_out_mm_out = MatMul (scaled_dot_product_attention_25, node_scaled_dot_product_attention_25_wo_t)
   [node_scaled_dot_product_attention_25_out_bias] node_scaled_dot_product_attention_25_out = Add (node_scaled_dot_product_attention_25_out_mm_out, "text.transformer.resblocks.25.attn.out_proj.bias")
   add_3519 = Add (add_3404, node_scaled_dot_product_attention_25_out)
   layer_norm_51 = LayerNormalization <axis: int = -1, epsilon: float = 1e-06, stash_type: int = 1> (add_3519, "text.transformer.resblocks.25.ln_2.weight", "text.transformer.resblocks.25.ln_2.bias")
   val_136 = MatMul (layer_norm_51, val_80)
   linear_102 = Add (val_136, "text.transformer.resblocks.25.mlp.c_fc.bias")
   gelu_25 = Gelu <approximate: string = "tanh"> (linear_102)
   val_137 = MatMul (gelu_25, val_81)
   linear_103 = Add (val_137, "text.transformer.resblocks.25.mlp.c_proj.bias")
   add_3540 = Add (add_3519, linear_103)
   layer_norm_52 = LayerNormalization <axis: int = -1, epsilon: float = 1e-06, stash_type: int = 1> (add_3540, "text.transformer.resblocks.26.ln_1.weight", "text.transformer.resblocks.26.ln_1.bias")
   [node_scaled_dot_product_attention_26_qkv_mm] node_scaled_dot_product_attention_26_qkv_mm_out = MatMul (layer_norm_52, val_82)
   [node_scaled_dot_product_attention_26_qkv_bias] node_scaled_dot_product_attention_26_qkv = Add (node_scaled_dot_product_attention_26_qkv_mm_out, "text.transformer.resblocks.26.attn.in_proj_bias")
   [node_scaled_dot_product_attention_26_qkv_split] node_scaled_dot_product_attention_26_q, node_scaled_dot_product_attention_26_k, node_scaled_dot_product_attention_26_v = Split <axis: int = -1> (node_scaled_dot_product_attention_26_qkv, attn3d_split_3x1152)
   [pool_hoist_node_scaled_dot_product_attention_26_q] node_scaled_dot_product_attention_26_q_pooled = Slice (node_scaled_dot_product_attention_26_q, val_0, val_3, val_2)
   scaled_dot_product_attention_26 = Attention <is_causal: int = 0, kv_num_heads: int = 16, q_num_heads: int = 16, qk_matmul_output_mode: int = 0, softcap: float = 0> (node_scaled_dot_product_attention_26_q_pooled, node_scaled_dot_product_attention_26_k, node_scaled_dot_product_attention_26_v)
   [node_scaled_dot_product_attention_26_out_mm] node_scaled_dot_product_attention_26_out_mm_out = MatMul (scaled_dot_product_attention_26, node_scaled_dot_product_attention_26_wo_t)
   [node_scaled_dot_product_attention_26_out_bias] node_scaled_dot_product_attention_26_out = Add (node_scaled_dot_product_attention_26_out_mm_out, "text.transformer.resblocks.26.attn.out_proj.bias")
   [pool_hoist_add_3540] add_3540_pooled = Slice (add_3540, val_0, val_3, val_2)
   add_3655 = Add (add_3540_pooled, node_scaled_dot_product_attention_26_out)
   layer_norm_53 = LayerNormalization <axis: int = -1, epsilon: float = 1e-06, stash_type: int = 1> (add_3655, "text.transformer.resblocks.26.ln_2.weight", "text.transformer.resblocks.26.ln_2.bias")
   val_138 = MatMul (layer_norm_53, val_83)
   linear_106 = Add (val_138, "text.transformer.resblocks.26.mlp.c_fc.bias")
   gelu_26 = Gelu <approximate: string = "tanh"> (linear_106)
   val_139 = MatMul (gelu_26, val_84)
   linear_107 = Add (val_139, "text.transformer.resblocks.26.mlp.c_proj.bias")
   add_3676 = Add (add_3655, linear_107)
   val_140 = Squeeze (add_3676, val_2)
   select_81 = LayerNormalization <axis: int = -1, epsilon: float = 1e-06, stash_type: int = 1> (val_140, "text.ln_final.weight", "text.ln_final.bias")
   linear_108 = Gemm <alpha: float = 1, beta: float = 1, transA: int = 0, transB: int = 1> (select_81, "text.text_projection.weight", "text.text_projection.bias")
   [node_linalg_vector_norm] linalg_vector_norm = ReduceL2 <keepdims: int = 1, noop_with_empty_axes: int = 0> (linear_108, val_0)
   [node_clamp_min] clamp_min = Clip (linalg_vector_norm, val_1)
   [node_div] text_embedding = Div (linear_108, clamp_min)
}

weights:
attn3d_split_3x1152 INT64[3] 125b254aeb28
node_scaled_dot_product_attention_10_wo_t FLOAT[1152,1152] 95bed049a51e
node_scaled_dot_product_attention_11_wo_t FLOAT[1152,1152] fab1f16c923f
node_scaled_dot_product_attention_12_wo_t FLOAT[1152,1152] 961e11dd5706
node_scaled_dot_product_attention_13_wo_t FLOAT[1152,1152] b66817f10a08
node_scaled_dot_product_attention_14_wo_t FLOAT[1152,1152] ec99b354dbbb
node_scaled_dot_product_attention_15_wo_t FLOAT[1152,1152] 520bd2b9d89f
node_scaled_dot_product_attention_16_wo_t FLOAT[1152,1152] fb06ce4799c4
node_scaled_dot_product_attention_17_wo_t FLOAT[1152,1152] e5cb91ac3f34
node_scaled_dot_product_attention_18_wo_t FLOAT[1152,1152] 9034271c0e02
node_scaled_dot_product_attention_19_wo_t FLOAT[1152,1152] dbaed71f883d
node_scaled_dot_product_attention_1_wo_t FLOAT[1152,1152] c91dbaf4d49d
node_scaled_dot_product_attention_20_wo_t FLOAT[1152,1152] 711825c7df92
node_scaled_dot_product_attention_21_wo_t FLOAT[1152,1152] 182517770504
node_scaled_dot_product_attention_22_wo_t FLOAT[1152,1152] 14d3cbd83f27
node_scaled_dot_product_attention_23_wo_t FLOAT[1152,1152] bcf4f260257e
node_scaled_dot_product_attention_24_wo_t FLOAT[1152,1152] d5d032cf4ed3
node_scaled_dot_product_attention_25_wo_t FLOAT[1152,1152] ae11622931b6
node_scaled_dot_product_attention_26_wo_t FLOAT[1152,1152] 35916aec54ca
node_scaled_dot_product_attention_2_wo_t FLOAT[1152,1152] 9c595f384802
node_scaled_dot_product_attention_3_wo_t FLOAT[1152,1152] e563a1878a6b
node_scaled_dot_product_attention_4_wo_t FLOAT[1152,1152] 3bad790a8904
node_scaled_dot_product_attention_5_wo_t FLOAT[1152,1152] 06ecb4e6d344
node_scaled_dot_product_attention_6_wo_t FLOAT[1152,1152] 4efb8d45b0c8
node_scaled_dot_product_attention_7_wo_t FLOAT[1152,1152] 893962ac2a6c
node_scaled_dot_product_attention_8_wo_t FLOAT[1152,1152] d0bb035473c3
node_scaled_dot_product_attention_9_wo_t FLOAT[1152,1152] 12f55c0a4ec2
node_scaled_dot_product_attention_wo_t FLOAT[1152,1152] ba057d75d576
text.ln_final.bias FLOAT[1152] fda5184bcd31
text.ln_final.weight FLOAT[1152] 1979cb6427f8
text.positional_embedding FLOAT[64,1152] 19959e1c0a8a
text.text_projection.bias FLOAT[1152] 4044de9e6e3f
text.text_projection.weight FLOAT[1152,1152] d58ab3e91b3e
text.token_embedding.weight_fp16 FLOAT16[256000,1152] 98d6c0a291d6
text.transformer.resblocks.0.attn.in_proj_bias FLOAT[3456] c7e76a0e88d2
text.transformer.resblocks.0.attn.out_proj.bias FLOAT[1152] 9a086c3bf404
text.transformer.resblocks.0.ln_1.bias FLOAT[1152] 6551bcfdcd13
text.transformer.resblocks.0.ln_1.weight FLOAT[1152] 17f0fa3f58e7
text.transformer.resblocks.0.ln_2.bias FLOAT[1152] f67884058a62
text.transformer.resblocks.0.ln_2.weight FLOAT[1152] 425130757884
text.transformer.resblocks.0.mlp.c_fc.bias FLOAT[4304] 96c465d064b3
text.transformer.resblocks.0.mlp.c_proj.bias FLOAT[1152] f472a00c03ce
text.transformer.resblocks.1.attn.in_proj_bias FLOAT[3456] 0af7fa7ea1d6
text.transformer.resblocks.1.attn.out_proj.bias FLOAT[1152] d9de85a4dae6
text.transformer.resblocks.1.ln_1.bias FLOAT[1152] 5f736fc75b9c
text.transformer.resblocks.1.ln_1.weight FLOAT[1152] 61c2d513903e
text.transformer.resblocks.1.ln_2.bias FLOAT[1152] 7b4d2261791b
text.transformer.resblocks.1.ln_2.weight FLOAT[1152] 1a339ee631ad
text.transformer.resblocks.1.mlp.c_fc.bias FLOAT[4304] 0e30061d9549
text.transformer.resblocks.1.mlp.c_proj.bias FLOAT[1152] ef31305e966e
text.transformer.resblocks.10.attn.in_proj_bias FLOAT[3456] 9804946d1b96
text.transformer.resblocks.10.attn.out_proj.bias FLOAT[1152] 2fe6c92cbe1f
text.transformer.resblocks.10.ln_1.bias FLOAT[1152] 081ad41827bc
text.transformer.resblocks.10.ln_1.weight FLOAT[1152] 24d0058a7b24
text.transformer.resblocks.10.ln_2.bias FLOAT[1152] 1e5f0575a56d
text.transformer.resblocks.10.ln_2.weight FLOAT[1152] 07e7edde6faf
text.transformer.resblocks.10.mlp.c_fc.bias FLOAT[4304] cce1a4be5d1d
text.transformer.resblocks.10.mlp.c_proj.bias FLOAT[1152] 210330571754
text.transformer.resblocks.11.attn.in_proj_bias FLOAT[3456] c148b7cf693f
text.transformer.resblocks.11.attn.out_proj.bias FLOAT[1152] 17dfde002e38
text.transformer.resblocks.11.ln_1.bias FLOAT[1152] ea709f83ede4
text.transformer.resblocks.11.ln_1.weight FLOAT[1152] 2d5c3007a307
text.transformer.resblocks.11.ln_2.bias FLOAT[1152] 9496ebdf2be7
text.transformer.resblocks.11.ln_2.weight FLOAT[1152] ae1d0704a7a8
text.transformer.resblocks.11.mlp.c_fc.bias FLOAT[4304] 4d52abc7cab3
text.transformer.resblocks.11.mlp.c_proj.bias FLOAT[1152] b34c8927b838
text.transformer.resblocks.12.attn.in_proj_bias FLOAT[3456] 9de78d054844
text.transformer.resblocks.12.attn.out_proj.bias FLOAT[1152] f2c1215727a7
text.transformer.resblocks.12.ln_1.bias FLOAT[1152] 0fa1dfbed91b
text.transformer.resblocks.12.ln_1.weight FLOAT[1152] 467891bccbc3
text.transformer.resblocks.12.ln_2.bias FLOAT[1152] b97206d296c3
text.transformer.resblocks.12.ln_2.weight FLOAT[1152] bf1fbf658324
text.transformer.resblocks.12.mlp.c_fc.bias FLOAT[4304] d297864f0069
text.transformer.resblocks.12.mlp.c_proj.bias FLOAT[1152] 14b24741eb30
text.transformer.resblocks.13.attn.in_proj_bias FLOAT[3456] 4bbf4fa5b06b
text.transformer.resblocks.13.attn.out_proj.bias FLOAT[1152] 71f1464fbade
text.transformer.resblocks.13.ln_1.bias FLOAT[1152] fd76b58e0138
text.transformer.resblocks.13.ln_1.weight FLOAT[1152] f98b9086037c
text.transformer.resblocks.13.ln_2.bias FLOAT[1152] 5395cd4d3633
text.transformer.resblocks.13.ln_2.weight FLOAT[1152] 40f6d1d4c71a
text.transformer.resblocks.13.mlp.c_fc.bias FLOAT[4304] 6f0ceb592e3a
text.transformer.resblocks.13.mlp.c_proj.bias FLOAT[1152] 7b49f148e471
text.transformer.resblocks.14.attn.in_proj_bias FLOAT[3456] 24847ffadd86
text.transformer.resblocks.14.attn.out_proj.bias FLOAT[1152] ed4d3c35d399
text.transformer.resblocks.14.ln_1.bias FLOAT[1152] 7b0897b7f969
text.transformer.resblocks.14.ln_1.weight FLOAT[1152] b61c8c653f3c
text.transformer.resblocks.14.ln_2.bias FLOAT[1152] d29a7ecd85f6
text.transformer.resblocks.14.ln_2.weight FLOAT[1152] 263c726762f1
text.transformer.resblocks.14.mlp.c_fc.bias FLOAT[4304] 70758c632f13
text.transformer.resblocks.14.mlp.c_proj.bias FLOAT[1152] 09b615abd1af
text.transformer.resblocks.15.attn.in_proj_bias FLOAT[3456] eb84c307c007
text.transformer.resblocks.15.attn.out_proj.bias FLOAT[1152] 9b0d9ef26785
text.transformer.resblocks.15.ln_1.bias FLOAT[1152] bf4eda1ddd7c
text.transformer.resblocks.15.ln_1.weight FLOAT[1152] d6babf55c7f5
text.transformer.resblocks.15.ln_2.bias FLOAT[1152] f71b1d5e3594
text.transformer.resblocks.15.ln_2.weight FLOAT[1152] 978be5d9c3cf
text.transformer.resblocks.15.mlp.c_fc.bias FLOAT[4304] 32550706c7f2
text.transformer.resblocks.15.mlp.c_proj.bias FLOAT[1152] ccb1c7c7ad6d
text.transformer.resblocks.16.attn.in_proj_bias FLOAT[3456] c12f1adefebc
text.transformer.resblocks.16.attn.out_proj.bias FLOAT[1152] 5d42f79c482b
text.transformer.resblocks.16.ln_1.bias FLOAT[1152] 7c844982240d
text.transformer.resblocks.16.ln_1.weight FLOAT[1152] 17c2178ad1f8
text.transformer.resblocks.16.ln_2.bias FLOAT[1152] 475dfbad0860
text.transformer.resblocks.16.ln_2.weight FLOAT[1152] 377192053f13
text.transformer.resblocks.16.mlp.c_fc.bias FLOAT[4304] a2507ef24887
text.transformer.resblocks.16.mlp.c_proj.bias FLOAT[1152] 4096a8235019
text.transformer.resblocks.17.attn.in_proj_bias FLOAT[3456] b69bf2a640ef
text.transformer.resblocks.17.attn.out_proj.bias FLOAT[1152] df1b057547ec
text.transformer.resblocks.17.ln_1.bias FLOAT[1152] d68cc72c33c6
text.transformer.resblocks.17.ln_1.weight FLOAT[1152] e213f89311e9
text.transformer.resblocks.17.ln_2.bias FLOAT[1152] c14066de06f9
text.transformer.resblocks.17.ln_2.weight FLOAT[1152] 29c12a15c55e
text.transformer.resblocks.17.mlp.c_fc.bias FLOAT[4304] 76750938f347
text.transformer.resblocks.17.mlp.c_proj.bias FLOAT[1152] 1f2947803ebf
text.transformer.resblocks.18.attn.in_proj_bias FLOAT[3456] 43d1cd18bdc7
text.transformer.resblocks.18.attn.out_proj.bias FLOAT[1152] b5041e00caa6
text.transformer.resblocks.18.ln_1.bias FLOAT[1152] 02d839e3d72c
text.transformer.resblocks.18.ln_1.weight FLOAT[1152] cd08db64d4ac
text.transformer.resblocks.18.ln_2.bias FLOAT[1152] af58591528d4
text.transformer.resblocks.18.ln_2.weight FLOAT[1152] 84129c330a21
text.transformer.resblocks.18.mlp.c_fc.bias FLOAT[4304] 8f308785b599
text.transformer.resblocks.18.mlp.c_proj.bias FLOAT[1152] c8ac4ddd5d44
text.transformer.resblocks.19.attn.in_proj_bias FLOAT[3456] e8b0dbfc64db
text.transformer.resblocks.19.attn.out_proj.bias FLOAT[1152] 3624fafbe663
text.transformer.resblocks.19.ln_1.bias FLOAT[1152] 92a0701066a8
text.transformer.resblocks.19.ln_1.weight FLOAT[1152] 8ef3dbb2a846
text.transformer.resblocks.19.ln_2.bias FLOAT[1152] 9c177af2ee9b
text.transformer.resblocks.19.ln_2.weight FLOAT[1152] b64580111d08
text.transformer.resblocks.19.mlp.c_fc.bias FLOAT[4304] 57bc3013cd4c
text.transformer.resblocks.19.mlp.c_proj.bias FLOAT[1152] 04a639a1c38c
text.transformer.resblocks.2.attn.in_proj_bias FLOAT[3456] cfdc50cc3a7a
text.transformer.resblocks.2.attn.out_proj.bias FLOAT[1152] 6ffd1d8ff308
text.transformer.resblocks.2.ln_1.bias FLOAT[1152] ecff997bfef9
text.transformer.resblocks.2.ln_1.weight FLOAT[1152] 76206e27dfeb
text.transformer.resblocks.2.ln_2.bias FLOAT[1152] 15dce61c7196
text.transformer.resblocks.2.ln_2.weight FLOAT[1152] fe167de8ed4c
text.transformer.resblocks.2.mlp.c_fc.bias FLOAT[4304] fe295ca97d1e
text.transformer.resblocks.2.mlp.c_proj.bias FLOAT[1152] 693aee5e5d98
text.transformer.resblocks.20.attn.in_proj_bias FLOAT[3456] a36fab5e2f98
text.transformer.resblocks.20.attn.out_proj.bias FLOAT[1152] ec6e90462ab2
text.transformer.resblocks.20.ln_1.bias FLOAT[1152] 7940bac76892
text.transformer.resblocks.20.ln_1.weight FLOAT[1152] f8a6a4a6cb9f
text.transformer.resblocks.20.ln_2.bias FLOAT[1152] 976b8a0e21cf
text.transformer.resblocks.20.ln_2.weight FLOAT[1152] 686b769528f0
text.transformer.resblocks.20.mlp.c_fc.bias FLOAT[4304] 4dd6190a2abc
text.transformer.resblocks.20.mlp.c_proj.bias FLOAT[1152] 38dc2cd89c00
text.transformer.resblocks.21.attn.in_proj_bias FLOAT[3456] 8354cd9975e7
text.transformer.resblocks.21.attn.out_proj.bias FLOAT[1152] cf3c2ae51ffa
text.transformer.resblocks.21.ln_1.bias FLOAT[1152] 3cca63d12226
text.transformer.resblocks.21.ln_1.weight FLOAT[1152] 8012ec83086a
text.transformer.resblocks.21.ln_2.bias FLOAT[1152] a576110049d3
text.transformer.resblocks.21.ln_2.weight FLOAT[1152] 087932f133da
text.transformer.resblocks.21.mlp.c_fc.bias FLOAT[4304] 7ea83f292466
text.transformer.resblocks.21.mlp.c_proj.bias FLOAT[1152] 552e0269a2dc
text.transformer.resblocks.22.attn.in_proj_bias FLOAT[3456] b8095e589e6b
text.transformer.resblocks.22.attn.out_proj.bias FLOAT[1152] 1eb9e169dc4a
text.transformer.resblocks.22.ln_1.bias FLOAT[1152] 688a8f797e08
text.transformer.resblocks.22.ln_1.weight FLOAT[1152] 3f3203144e4d
text.transformer.resblocks.22.ln_2.bias FLOAT[1152] 84aa20ce06e9
text.transformer.resblocks.22.ln_2.weight FLOAT[1152] 65916739a7ab
text.transformer.resblocks.22.mlp.c_fc.bias FLOAT[4304] 6dddca9eed17
text.transformer.resblocks.22.mlp.c_proj.bias FLOAT[1152] 24a4c5a75dde
text.transformer.resblocks.23.attn.in_proj_bias FLOAT[3456] 3738665c9f66
text.transformer.resblocks.23.attn.out_proj.bias FLOAT[1152] 6c170a06f346
text.transformer.resblocks.23.ln_1.bias FLOAT[1152] 9ce54f67873e
text.transformer.resblocks.23.ln_1.weight FLOAT[1152] 6e0268bba804
text.transformer.resblocks.23.ln_2.bias FLOAT[1152] 9c3f29e90fca
text.transformer.resblocks.23.ln_2.weight FLOAT[1152] 919e3433888e
text.transformer.resblocks.23.mlp.c_fc.bias FLOAT[4304] 36bfc2371b66
text.transformer.resblocks.23.mlp.c_proj.bias FLOAT[1152] 9505cf901b5e
text.transformer.resblocks.24.attn.in_proj_bias FLOAT[3456] e66c90229023
text.transformer.resblocks.24.attn.out_proj.bias FLOAT[1152] de5f89770a8b
text.transformer.resblocks.24.ln_1.bias FLOAT[1152] fa4d27bfc39d
text.transformer.resblocks.24.ln_1.weight FLOAT[1152] 582d37c1aab5
text.transformer.resblocks.24.ln_2.bias FLOAT[1152] 4e4c1d3bf95e
text.transformer.resblocks.24.ln_2.weight FLOAT[1152] 9dc9add0d51f
text.transformer.resblocks.24.mlp.c_fc.bias FLOAT[4304] badf449a5579
text.transformer.resblocks.24.mlp.c_proj.bias FLOAT[1152] 593ac9fe46f3
text.transformer.resblocks.25.attn.in_proj_bias FLOAT[3456] 4c55e1c02ecd
text.transformer.resblocks.25.attn.out_proj.bias FLOAT[1152] e8c7f1ea2be5
text.transformer.resblocks.25.ln_1.bias FLOAT[1152] 683fc7d9028b
text.transformer.resblocks.25.ln_1.weight FLOAT[1152] de7e01703fef
text.transformer.resblocks.25.ln_2.bias FLOAT[1152] 30a58af3cf94
text.transformer.resblocks.25.ln_2.weight FLOAT[1152] a52c1ef9e03b
text.transformer.resblocks.25.mlp.c_fc.bias FLOAT[4304] 45ce68ab56da
text.transformer.resblocks.25.mlp.c_proj.bias FLOAT[1152] 821af6945deb
text.transformer.resblocks.26.attn.in_proj_bias FLOAT[3456] 508085bdc90c
text.transformer.resblocks.26.attn.out_proj.bias FLOAT[1152] 9b0118d3bbea
text.transformer.resblocks.26.ln_1.bias FLOAT[1152] ef2a760667b3
text.transformer.resblocks.26.ln_1.weight FLOAT[1152] c5b8a234eafb
text.transformer.resblocks.26.ln_2.bias FLOAT[1152] a3d7f6ca935b
text.transformer.resblocks.26.ln_2.weight FLOAT[1152] 1590fdf6cf5b
text.transformer.resblocks.26.mlp.c_fc.bias FLOAT[4304] cb19c32e3d82
text.transformer.resblocks.26.mlp.c_proj.bias FLOAT[1152] 699fd3b87c38
text.transformer.resblocks.3.attn.in_proj_bias FLOAT[3456] 0b63818e3812
text.transformer.resblocks.3.attn.out_proj.bias FLOAT[1152] dcc802734671
text.transformer.resblocks.3.ln_1.bias FLOAT[1152] 93ae42fea119
text.transformer.resblocks.3.ln_1.weight FLOAT[1152] 29735949357a
text.transformer.resblocks.3.ln_2.bias FLOAT[1152] 869977543c8f
text.transformer.resblocks.3.ln_2.weight FLOAT[1152] 0395fc2bbdb5
text.transformer.resblocks.3.mlp.c_fc.bias FLOAT[4304] 572b109bf72f
text.transformer.resblocks.3.mlp.c_proj.bias FLOAT[1152] 5f614e7a462b
text.transformer.resblocks.4.attn.in_proj_bias FLOAT[3456] 5cbe376e510f
text.transformer.resblocks.4.attn.out_proj.bias FLOAT[1152] e26ab7203196
text.transformer.resblocks.4.ln_1.bias FLOAT[1152] 06ca1eceabd9
text.transformer.resblocks.4.ln_1.weight FLOAT[1152] 034b86e43a39
text.transformer.resblocks.4.ln_2.bias FLOAT[1152] b107e9136eb7
text.transformer.resblocks.4.ln_2.weight FLOAT[1152] 954ba4b59a82
text.transformer.resblocks.4.mlp.c_fc.bias FLOAT[4304] 14a32ce02daf
text.transformer.resblocks.4.mlp.c_proj.bias FLOAT[1152] ea994f89c8ef
text.transformer.resblocks.5.attn.in_proj_bias FLOAT[3456] 401713e4cd3f
text.transformer.resblocks.5.attn.out_proj.bias FLOAT[1152] 9d7f67d9d422
text.transformer.resblocks.5.ln_1.bias FLOAT[1152] ca9b542ab4fa
text.transformer.resblocks.5.ln_1.weight FLOAT[1152] 2e4cd3bb2a83
text.transformer.resblocks.5.ln_2.bias FLOAT[1152] 6fb64c7fd6d5
text.transformer.resblocks.5.ln_2.weight FLOAT[1152] c55962520e03
text.transformer.resblocks.5.mlp.c_fc.bias FLOAT[4304] 2c6988f8e3d4
text.transformer.resblocks.5.mlp.c_proj.bias FLOAT[1152] 322a51c56b1c
text.transformer.resblocks.6.attn.in_proj_bias FLOAT[3456] ba8e2c432c3a
text.transformer.resblocks.6.attn.out_proj.bias FLOAT[1152] 9636b598e89a
text.transformer.resblocks.6.ln_1.bias FLOAT[1152] 4f580d1c7577
text.transformer.resblocks.6.ln_1.weight FLOAT[1152] c320eb873847
text.transformer.resblocks.6.ln_2.bias FLOAT[1152] bd3bab6b61cc
text.transformer.resblocks.6.ln_2.weight FLOAT[1152] 58bec68beb68
text.transformer.resblocks.6.mlp.c_fc.bias FLOAT[4304] 62dce5c474b1
text.transformer.resblocks.6.mlp.c_proj.bias FLOAT[1152] d649e3e8f070
text.transformer.resblocks.7.attn.in_proj_bias FLOAT[3456] 3fa22542a7c3
text.transformer.resblocks.7.attn.out_proj.bias FLOAT[1152] 9803f0fbe8c7
text.transformer.resblocks.7.ln_1.bias FLOAT[1152] 8ab9e1b8a9b7
text.transformer.resblocks.7.ln_1.weight FLOAT[1152] 24c8496f758f
text.transformer.resblocks.7.ln_2.bias FLOAT[1152] 043c7112abcd
text.transformer.resblocks.7.ln_2.weight FLOAT[1152] 1c137f781abf
text.transformer.resblocks.7.mlp.c_fc.bias FLOAT[4304] 917176b8cc61
text.transformer.resblocks.7.mlp.c_proj.bias FLOAT[1152] b8d2e5d7c203
text.transformer.resblocks.8.attn.in_proj_bias FLOAT[3456] 765d77a1b393
text.transformer.resblocks.8.attn.out_proj.bias FLOAT[1152] 4c58100ec455
text.transformer.resblocks.8.ln_1.bias FLOAT[1152] 1c5869458be4
text.transformer.resblocks.8.ln_1.weight FLOAT[1152] c878d46e55d8
text.transformer.resblocks.8.ln_2.bias FLOAT[1152] 5d0ac04ac51f
text.transformer.resblocks.8.ln_2.weight FLOAT[1152] b1b0371590a4
text.transformer.resblocks.8.mlp.c_fc.bias FLOAT[4304] 68d5b359a367
text.transformer.resblocks.8.mlp.c_proj.bias FLOAT[1152] e90797433818
text.transformer.resblocks.9.attn.in_proj_bias FLOAT[3456] 31b9129b9681
text.transformer.resblocks.9.attn.out_proj.bias FLOAT[1152] 9c9b3f182663
text.transformer.resblocks.9.ln_1.bias FLOAT[1152] 2f8ff7ecb908
text.transformer.resblocks.9.ln_1.weight FLOAT[1152] 8489cfdd2377
text.transformer.resblocks.9.ln_2.bias FLOAT[1152] 32705226a93c
text.transformer.resblocks.9.ln_2.weight FLOAT[1152] d99afb1c5e19
text.transformer.resblocks.9.mlp.c_fc.bias FLOAT[4304] 230440e97505
text.transformer.resblocks.9.mlp.c_proj.bias FLOAT[1152] 8c8f2ed00757
val_0 INT64[1] 12a3ae445661
val_1 FLOAT[] 6708d9be4956
val_10 FLOAT[1152,3456] f7f5edd0a84f
val_11 FLOAT[1152,4304] 828d9c45c62b
val_12 FLOAT[4304,1152] e070c8678238
val_13 FLOAT[1152,3456] 230859fdcd04
val_14 FLOAT[1152,4304] 57cf0c97c751
val_15 FLOAT[4304,1152] 8665d58ce942
val_16 FLOAT[1152,3456] e21c5515e9bf
val_17 FLOAT[1152,4304] 14256da0b922
val_18 FLOAT[4304,1152] eb78bebb357d
val_19 FLOAT[1152,3456] 5dafae4ae8ae
val_2 INT64[1] 7c9fa136d441
val_20 FLOAT[1152,4304] b1f68f6633b6
val_21 FLOAT[4304,1152] 7ce74fa5e7be
val_22 FLOAT[1152,3456] 5e7cf5b3e4d9
val_23 FLOAT[1152,4304] e23936140c2f
val_24 FLOAT[4304,1152] 5deb5e3600b9
val_25 FLOAT[1152,3456] e0e59adccf72
val_26 FLOAT[1152,4304] 3a867c061a9b
val_27 FLOAT[4304,1152] 8ade918276e2
val_28 FLOAT[1152,3456] 7b6a0fcf916c
val_29 FLOAT[1152,4304] 3969284ce9f3
val_3 INT64[1] 6a69a6cc7473
val_30 FLOAT[4304,1152] 36bc635aac73
val_31 FLOAT[1152,3456] c6492057ea99
val_32 FLOAT[1152,4304] dfcce34b5248
val_33 FLOAT[4304,1152] 122144a6c3de
val_34 FLOAT[1152,3456] d283a54ab2ed
val_35 FLOAT[1152,4304] 3b4f5502ffa4
val_36 FLOAT[4304,1152] 9dbc6acde410
val_37 FLOAT[1152,3456] ffd4f7014886
val_38 FLOAT[1152,4304] 6bc04e145132
val_39 FLOAT[4304,1152] 179c77ba5d5c
val_4 FLOAT[1152,3456] dcf476e8a073
val_40 FLOAT[1152,3456] a50e25688eeb
val_41 FLOAT[1152,4304] 989ac7a70947
val_42 FLOAT[4304,1152] 576454028d42
val_43 FLOAT[1152,3456] 3d82bb3fa4cc
val_44 FLOAT[1152,4304] 66a4d57704ba
val_45 FLOAT[4304,1152] 87bfc7a41f7f
val_46 FLOAT[1152,3456] 76ec9a658630
val_47 FLOAT[1152,4304] deb625b77fd2
val_48 FLOAT[4304,1152] f4956afa37d4
val_49 FLOAT[1152,3456] 305ded262ef7
val_5 FLOAT[1152,4304] eb908399bebd
val_50 FLOAT[1152,4304] bad244b64174
val_51 FLOAT[4304,1152] 825081a9c34c
val_52 FLOAT[1152,3456] eb2d3b5398a4
val_53 FLOAT[1152,4304] c1177f749bfc
val_54 FLOAT[4304,1152] 6687b3089ad3
val_55 FLOAT[1152,3456] 72ed70d4ad5d
val_56 FLOAT[1152,4304] e8db6d25d89b
val_57 FLOAT[4304,1152] 58663607dab7
val_58 FLOAT[1152,3456] d91fd0bea361
val_59 FLOAT[1152,4304] fed86b142071
val_6 FLOAT[4304,1152] d764db0dd20d
val_60 FLOAT[4304,1152] 6a6a1f46c321
val_61 FLOAT[1152,3456] a9e1fbee8c65
val_62 FLOAT[1152,4304] 09e246b042f8
val_63 FLOAT[4304,1152] e878174d72de
val_64 FLOAT[1152,3456] 2b8a91f3eebc
val_65 FLOAT[1152,4304] 538f6ae487e4
val_66 FLOAT[4304,1152] 5ba0db2f18fa
val_67 FLOAT[1152,3456] 0bdfa16e8942
val_68 FLOAT[1152,4304] 30b40fae5694
val_69 FLOAT[4304,1152] 60722ab20917
val_7 FLOAT[1152,3456] 9ff47c0a6d74
val_70 FLOAT[1152,3456] d3d74f1b4ebb
val_71 FLOAT[1152,4304] 223061c6e998
val_72 FLOAT[4304,1152] acba0342796f
val_73 FLOAT[1152,3456] dfa1cc59c0c4
val_74 FLOAT[1152,4304] 12d8a0c2be19
val_75 FLOAT[4304,1152] b66bec20b48a
val_76 FLOAT[1152,3456] c89a4f247065
val_77 FLOAT[1152,4304] d3dfc2e14607
val_78 FLOAT[4304,1152] e154b3b7222a
val_79 FLOAT[1152,3456] 6399d2d1b79e
val_8 FLOAT[1152,4304] c49f0c3a65ce
val_80 FLOAT[1152,4304] 3e0b4c4e283f
val_81 FLOAT[4304,1152] f4b3b5dd790a
val_82 FLOAT[1152,3456] 8b5e02a1ebfe
val_83 FLOAT[1152,4304] 5e6b77b70637
val_84 FLOAT[4304,1152] 02319a351534
val_9 FLOAT[4304,1152] 2fa91a873c33
