Upload TRT model for Nvidia H200

#16
.gitattributes CHANGED
@@ -42,3 +42,4 @@ model_rtx_pro_6000_bf16.plan filter=lfs diff=lfs merge=lfs -text
42
  model_rtx_pro_6000_fp8.plan filter=lfs diff=lfs merge=lfs -text
43
  model_l40s_fp8.plan filter=lfs diff=lfs merge=lfs -text
44
  model_fp8.onnx.data filter=lfs diff=lfs merge=lfs -text
 
 
42
  model_rtx_pro_6000_fp8.plan filter=lfs diff=lfs merge=lfs -text
43
  model_l40s_fp8.plan filter=lfs diff=lfs merge=lfs -text
44
  model_fp8.onnx.data filter=lfs diff=lfs merge=lfs -text
45
+ model_h200_fp8.plan filter=lfs diff=lfs merge=lfs -text
model_h200_fp8.plan ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4b75a5c24d9ad5d89919c01ff90f7717f405f1cb578e1304b0ca02969665f535
3
+ size 1913984356
trt_engine_layer_summary_h200_fp8.txt ADDED
@@ -0,0 +1,602 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---- Resolved TRT Profile ----
2
+ MIN_BATCH=1
3
+ OPT_BATCH=3
4
+ MAX_BATCH=12
5
+ MIN_SEQ_LEN=1
6
+ OPT_SEQ_LEN=512
7
+ MAX_SEQ_LEN=512
8
+ WORKSPACE_SIZE=24696061952
9
+ BUILDER_OPTIMIZATION_LEVEL=3
10
+ PRECISION=fp8
11
+
12
+ ==== TensorRT Engine ====
13
+ Total Layers: 568
14
+
15
+ ==== Precision Statistics ====
16
+ Float: 1453
17
+ FP8: 366
18
+ Int64: 10
19
+ Int32: 6
20
+ Bool: 2
21
+
22
+ ==== Layer Type Statistics ====
23
+ kgen: 322
24
+ gemm: 235
25
+ TrainStation: 3
26
+ shape_call: 2
27
+ reshape: 2
28
+ cast: 2
29
+ NonZero: 1
30
+ DeviceToShapeHost: 1
31
+
32
+ ==== Layer List ====
33
+ # Type Precision Name
34
+ ----------------------------------------------------------------------------------------------------
35
+ 0 TrainStation - [trainStation1]
36
+ 1 shape_call - __mye505_0_myl1_0
37
+ 2 kgen Bool __myl_EqlReshSlicRepl_myl1_1
38
+ 3 NonZero Int32 /text_model/embed_tokens/NonZero
39
+ 4 DeviceToShapeHost - /text_model/embed_tokens/NonZero[size][DevicetoShapeHostCopy]
40
+ 5 TrainStation - [trainStation2]
41
+ 6 shape_call - __mye159904_0_myl5_0
42
+ 7 kgen Float __myl_IotaCastReshCast_myl5_1
43
+ 8 kgen Float __myl_MoveMul_myl5_2
44
+ 9 kgen Float __myl_CastGathMul_myl5_3
45
+ 10 kgen Int32 __myl_CastTranCast_myl5_4
46
+ 11 kgen Float __myl_Slic_myl5_5
47
+ 12 kgen Float __myl_Scat_myl5_6
48
+ 13 kgen FP8 __myl_MulMeanAddSqrtDivMulMulCastMulCastMulCast_myl5_7
49
+ 14 gemm Float /text_model/layers_0/self_attn/v_proj/MatMul_myl5_8
50
+ 15 kgen Float __myl_SlicRepl_myl5_9
51
+ 16 gemm Float /text_model/layers_0/self_attn/k_proj/MatMul_myl5_10
52
+ 17 gemm Float /text_model/layers_0/self_attn/q_proj/MatMul_myl5_11
53
+ 18 kgen Float __myl_TranMulMean_myl5_12
54
+ 19 kgen Float __myl_MulMeanAddSqrt_myl5_13
55
+ 20 kgen Float __myl_ConcSinCosReshReshAddSqrtDivMulDivMulMulSlicNegSlicConcMulMulAddReshSlicReplReshTranMulSlicEtc_myl5_14
56
+ 21 gemm Float /text_model/layers_0/self_attn/MatMul_myl5_15
57
+ 22 kgen Float __myl_CastSubMulSlicReplSlicReshAddMaxrSubExpSumDivMul_myl5_16
58
+ 23 kgen Float __myl_MoveReplConc_myl5_17
59
+ 24 kgen Float __myl_MoveReplConc_myl5_18
60
+ 25 gemm Float /text_model/layers_0/self_attn/MatMul_1_myl5_19
61
+ 26 kgen FP8 __myl_MulCast_myl5_20
62
+ 27 gemm Float /text_model/layers_0/self_attn/o_proj/MatMul_myl5_21
63
+ 28 kgen Float __myl_MulMeanAddSqrt_myl5_22
64
+ 29 kgen Float __myl_DivMulMulReshTranReshAddReshMulMeanAddSqrtDivMulMulCastMulCast_myl5_23
65
+ 30 gemm Float /text_model/layers_0/mlp/up_proj/MatMul_myl5_24
66
+ 31 gemm Float /text_model/layers_0/mlp/gate_proj/MatMul_myl5_25
67
+ 32 kgen FP8 __myl_MulMulMulAddMulTanhAddMulMulMulMulCast_myl5_26
68
+ 33 gemm Float /text_model/layers_0/mlp/down_proj/MatMul_myl5_27
69
+ 34 kgen Float __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulMulCastMulCastMulCast_myl5_28
70
+ 35 gemm Float /text_model/layers_1/self_attn/v_proj/MatMul_myl5_29
71
+ 36 kgen Float __myl_SlicRepl_myl5_30
72
+ 37 gemm Float /text_model/layers_1/self_attn/k_proj/MatMul_myl5_31
73
+ 38 gemm Float /text_model/layers_1/self_attn/q_proj/MatMul_myl5_32
74
+ 39 kgen Float __myl_MulMeanAddSqrt_myl5_33
75
+ 40 kgen Float __myl_DivMulMulSlicNegSlicConcMulMulAddReshSlicReplReshTran_myl5_34
76
+ 41 kgen Float __myl_TranMulMean_myl5_35
77
+ 42 kgen Float __myl_AddSqrtDivMulMulSlicNegSlicConcMulMulAdd_myl5_36
78
+ 43 gemm Float /text_model/layers_1/self_attn/MatMul_myl5_37
79
+ 44 kgen Float __myl_SlicReshAddMaxrSubExpSumDivMul_myl5_38
80
+ 45 kgen Float __myl_MoveReplConc_myl5_39
81
+ 46 kgen Float __myl_MoveReplConc_myl5_40
82
+ 47 gemm Float /text_model/layers_1/self_attn/MatMul_1_myl5_41
83
+ 48 kgen FP8 __myl_MulCast_myl5_42
84
+ 49 gemm Float /text_model/layers_1/self_attn/o_proj/MatMul_myl5_43
85
+ 50 kgen Float __myl_MulMeanAddSqrtDivMulMulReshTranReshAddMulMeanAddSqrtDivMulMulCastMulCast_myl5_44
86
+ 51 gemm Float /text_model/layers_1/mlp/up_proj/MatMul_myl5_45
87
+ 52 gemm Float /text_model/layers_1/mlp/gate_proj/MatMul_myl5_46
88
+ 53 kgen FP8 __myl_MulMulMulAddMulTanhAddMulMulMulMulCast_myl5_47
89
+ 54 gemm Float /text_model/layers_1/mlp/down_proj/MatMul_myl5_48
90
+ 55 kgen Float __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulMulCastMulCastMulCast_myl5_49
91
+ 56 gemm Float /text_model/layers_2/self_attn/v_proj/MatMul_myl5_50
92
+ 57 kgen Float __myl_TranReshReshSlicRepl_myl5_51
93
+ 58 gemm Float /text_model/layers_2/self_attn/k_proj/MatMul_myl5_52
94
+ 59 gemm Float /text_model/layers_2/self_attn/q_proj/MatMul_myl5_53
95
+ 60 kgen Float __myl_TranReshReshMulMean_myl5_54
96
+ 61 kgen Float __myl_AddSqrtDivMulMulSlicNegSlicConcMulMulAddReshSlicReplReshTran_myl5_55
97
+ 62 kgen Float __myl_TranReshTranMulMean_myl5_56
98
+ 63 kgen Float __myl_AddSqrtDivMulMulSlicNegSlicConcMulMulAdd_myl5_57
99
+ 64 gemm Float /text_model/layers_2/self_attn/MatMul_myl5_58
100
+ 65 kgen Float __myl_SlicReshAddMaxrSubExpSumDivMul_myl5_59
101
+ 66 kgen Float __myl_MoveReplConc_myl5_60
102
+ 67 kgen Float __myl_MoveReplConc_myl5_61
103
+ 68 gemm Float /text_model/layers_2/self_attn/MatMul_1_myl5_62
104
+ 69 kgen FP8 __myl_MulCast_myl5_63
105
+ 70 gemm Float /text_model/layers_2/self_attn/o_proj/MatMul_myl5_64
106
+ 71 kgen Float __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulMulCastMulCast_myl5_65
107
+ 72 gemm Float /text_model/layers_2/mlp/up_proj/MatMul_myl5_66
108
+ 73 gemm Float /text_model/layers_2/mlp/gate_proj/MatMul_myl5_67
109
+ 74 kgen FP8 __myl_MulMulMulAddMulTanhAddMulMulMulMulCast_myl5_68
110
+ 75 gemm Float /text_model/layers_2/mlp/down_proj/MatMul_myl5_69
111
+ 76 kgen Float __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulMulCastMulCastMulCast_myl5_70
112
+ 77 gemm Float /text_model/layers_3/self_attn/v_proj/MatMul_myl5_71
113
+ 78 kgen Float __myl_TranReshReshSlicRepl_myl5_72
114
+ 79 gemm Float /text_model/layers_3/self_attn/k_proj/MatMul_myl5_73
115
+ 80 gemm Float /text_model/layers_3/self_attn/q_proj/MatMul_myl5_74
116
+ 81 kgen Float __myl_TranReshReshMulMean_myl5_75
117
+ 82 kgen Float __myl_AddSqrtDivMulMulSlicNegSlicConcMulMulAddReshSlicReplReshTran_myl5_76
118
+ 83 kgen Float __myl_TranReshTranMulMean_myl5_77
119
+ 84 kgen Float __myl_AddSqrtDivMulMulSlicNegSlicConcMulMulAdd_myl5_78
120
+ 85 gemm Float /text_model/layers_3/self_attn/MatMul_myl5_79
121
+ 86 kgen Float __myl_SlicReshAddMaxrSubExpSumDivMul_myl5_80
122
+ 87 kgen Float __myl_MoveReplConc_myl5_81
123
+ 88 kgen Float __myl_MoveReplConc_myl5_82
124
+ 89 gemm Float /text_model/layers_3/self_attn/MatMul_1_myl5_83
125
+ 90 kgen FP8 __myl_MulCast_myl5_84
126
+ 91 gemm Float /text_model/layers_3/self_attn/o_proj/MatMul_myl5_85
127
+ 92 kgen Float __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulMulCastMulCast_myl5_86
128
+ 93 gemm Float /text_model/layers_3/mlp/up_proj/MatMul_myl5_87
129
+ 94 gemm Float /text_model/layers_3/mlp/gate_proj/MatMul_myl5_88
130
+ 95 kgen FP8 __myl_MulMulMulAddMulTanhAddMulMulMulMulCast_myl5_89
131
+ 96 gemm Float /text_model/layers_3/mlp/down_proj/MatMul_myl5_90
132
+ 97 kgen Float __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulMulCastMulCastMulCast_myl5_91
133
+ 98 gemm Float /text_model/layers_4/self_attn/v_proj/MatMul_myl5_92
134
+ 99 kgen Float __myl_TranReshReshSlicRepl_myl5_93
135
+ 100 gemm Float /text_model/layers_4/self_attn/k_proj/MatMul_myl5_94
136
+ 101 gemm Float /text_model/layers_4/self_attn/q_proj/MatMul_myl5_95
137
+ 102 kgen Float __myl_TranReshReshMulMean_myl5_96
138
+ 103 kgen Float __myl_AddSqrtDivMulMulSlicNegSlicConcMulMulAddReshSlicReplReshTran_myl5_97
139
+ 104 kgen Float __myl_TranReshTranMulMean_myl5_98
140
+ 105 kgen Float __myl_AddSqrtDivMulMulSlicNegSlicConcMulMulAdd_myl5_99
141
+ 106 gemm Float /text_model/layers_4/self_attn/MatMul_myl5_100
142
+ 107 kgen Float __myl_SlicReshAddMaxrSubExpSumDivMul_myl5_101
143
+ 108 kgen Float __myl_MoveReplConc_myl5_102
144
+ 109 kgen Float __myl_MoveReplConc_myl5_103
145
+ 110 gemm Float /text_model/layers_4/self_attn/MatMul_1_myl5_104
146
+ 111 kgen FP8 __myl_MulCast_myl5_105
147
+ 112 gemm Float /text_model/layers_4/self_attn/o_proj/MatMul_myl5_106
148
+ 113 kgen Float __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulMulCastMulCast_myl5_107
149
+ 114 gemm Float /text_model/layers_4/mlp/up_proj/MatMul_myl5_108
150
+ 115 gemm Float /text_model/layers_4/mlp/gate_proj/MatMul_myl5_109
151
+ 116 kgen FP8 __myl_MulMulMulAddMulTanhAddMulMulMulMulCast_myl5_110
152
+ 117 gemm Float /text_model/layers_4/mlp/down_proj/MatMul_myl5_111
153
+ 118 kgen Float __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulMulCastMulCastMulCast_myl5_112
154
+ 119 gemm Float /text_model/layers_5/self_attn/v_proj/MatMul_myl5_113
155
+ 120 kgen Float __myl_TranReshReshSlicRepl_myl5_114
156
+ 121 gemm Float /text_model/layers_5/self_attn/k_proj/MatMul_myl5_115
157
+ 122 gemm Float /text_model/layers_5/self_attn/q_proj/MatMul_myl5_116
158
+ 123 kgen Float __myl_TranReshReshMulMean_myl5_117
159
+ 124 kgen Float __myl_TranReshTranMulMean_myl5_118
160
+ 125 reshape Int64 __mye578893_myl5_119
161
+ 126 cast Int64 cast_hvar^578889i64_myl5_120
162
+ 127 reshape Int64 __mye578919_myl5_121
163
+ 128 cast Int64 cast_hvar^578915i64_myl5_122
164
+ 129 kgen Float __myl_ConcSinCosReshReshAddSqrtAddSqrtDivMulDivMulMulSlicNegSlicConcMulMulAddReshSlicReplReshTranEtc_myl5_123
165
+ 130 gemm Float /text_model/layers_5/self_attn/MatMul_myl5_124
166
+ 131 kgen Float __myl_SlicReshAddMaxrSubExpSumDivMul_myl5_125
167
+ 132 kgen Float __myl_MoveReplConc_myl5_126
168
+ 133 kgen Float __myl_MoveReplConc_myl5_127
169
+ 134 gemm Float /text_model/layers_5/self_attn/MatMul_1_myl5_128
170
+ 135 kgen FP8 __myl_MulCast_myl5_129
171
+ 136 gemm Float /text_model/layers_5/self_attn/o_proj/MatMul_myl5_130
172
+ 137 kgen Float __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulMulCastMulCast_myl5_131
173
+ 138 gemm Float /text_model/layers_5/mlp/up_proj/MatMul_myl5_132
174
+ 139 gemm Float /text_model/layers_5/mlp/gate_proj/MatMul_myl5_133
175
+ 140 kgen FP8 __myl_MulMulMulAddMulTanhAddMulMulMulMulCast_myl5_134
176
+ 141 gemm Float /text_model/layers_5/mlp/down_proj/MatMul_myl5_135
177
+ 142 kgen Float __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulMulCastMulCastMulCast_myl5_136
178
+ 143 gemm Float /text_model/layers_6/self_attn/v_proj/MatMul_myl5_137
179
+ 144 kgen Float __myl_TranReshReshSlicRepl_myl5_138
180
+ 145 gemm Float /text_model/layers_6/self_attn/k_proj/MatMul_myl5_139
181
+ 146 gemm Float /text_model/layers_6/self_attn/q_proj/MatMul_myl5_140
182
+ 147 kgen Float __myl_TranReshReshMulMean_myl5_141
183
+ 148 kgen Float __myl_AddSqrtDivMulMulSlicNegSlicConcMulMulAddReshSlicReplReshTran_myl5_142
184
+ 149 kgen Float __myl_TranReshTranMulMean_myl5_143
185
+ 150 kgen Float __myl_AddSqrtDivMulMulSlicNegSlicConcMulMulAdd_myl5_144
186
+ 151 gemm Float /text_model/layers_6/self_attn/MatMul_myl5_145
187
+ 152 kgen Float __myl_SlicReshAddMaxrSubExpSumDivMul_myl5_146
188
+ 153 kgen Float __myl_MoveReplConc_myl5_147
189
+ 154 kgen Float __myl_MoveReplConc_myl5_148
190
+ 155 gemm Float /text_model/layers_6/self_attn/MatMul_1_myl5_149
191
+ 156 kgen FP8 __myl_MulCast_myl5_150
192
+ 157 gemm Float /text_model/layers_6/self_attn/o_proj/MatMul_myl5_151
193
+ 158 kgen Float __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulMulCastMulCast_myl5_152
194
+ 159 gemm Float /text_model/layers_6/mlp/up_proj/MatMul_myl5_153
195
+ 160 gemm Float /text_model/layers_6/mlp/gate_proj/MatMul_myl5_154
196
+ 161 kgen FP8 __myl_MulMulMulAddMulTanhAddMulMulMulMulCast_myl5_155
197
+ 162 gemm Float /text_model/layers_6/mlp/down_proj/MatMul_myl5_156
198
+ 163 kgen Float __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulMulCastMulCastMulCast_myl5_157
199
+ 164 gemm Float /text_model/layers_7/self_attn/v_proj/MatMul_myl5_158
200
+ 165 kgen Float __myl_TranReshReshSlicRepl_myl5_159
201
+ 166 gemm Float /text_model/layers_7/self_attn/k_proj/MatMul_myl5_160
202
+ 167 gemm Float /text_model/layers_7/self_attn/q_proj/MatMul_myl5_161
203
+ 168 kgen Float __myl_TranReshReshMulMean_myl5_162
204
+ 169 kgen Float __myl_AddSqrtDivMulMulSlicNegSlicConcMulMulAddReshSlicReplReshTran_myl5_163
205
+ 170 kgen Float __myl_TranReshTranMulMean_myl5_164
206
+ 171 kgen Float __myl_AddSqrtDivMulMulSlicNegSlicConcMulMulAdd_myl5_165
207
+ 172 gemm Float /text_model/layers_7/self_attn/MatMul_myl5_166
208
+ 173 kgen Float __myl_SlicReshAddMaxrSubExpSumDivMul_myl5_167
209
+ 174 kgen Float __myl_MoveReplConc_myl5_168
210
+ 175 kgen Float __myl_MoveReplConc_myl5_169
211
+ 176 gemm Float /text_model/layers_7/self_attn/MatMul_1_myl5_170
212
+ 177 kgen FP8 __myl_MulCast_myl5_171
213
+ 178 gemm Float /text_model/layers_7/self_attn/o_proj/MatMul_myl5_172
214
+ 179 kgen Float __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulMulCastMulCast_myl5_173
215
+ 180 gemm Float /text_model/layers_7/mlp/up_proj/MatMul_myl5_174
216
+ 181 gemm Float /text_model/layers_7/mlp/gate_proj/MatMul_myl5_175
217
+ 182 kgen FP8 __myl_MulMulMulAddMulTanhAddMulMulMulMulCast_myl5_176
218
+ 183 gemm Float /text_model/layers_7/mlp/down_proj/MatMul_myl5_177
219
+ 184 kgen Float __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulMulCastMulCastMulCast_myl5_178
220
+ 185 gemm Float /text_model/layers_8/self_attn/v_proj/MatMul_myl5_179
221
+ 186 kgen Float __myl_TranReshReshSlicRepl_myl5_180
222
+ 187 gemm Float /text_model/layers_8/self_attn/k_proj/MatMul_myl5_181
223
+ 188 gemm Float /text_model/layers_8/self_attn/q_proj/MatMul_myl5_182
224
+ 189 kgen Float __myl_TranReshReshMulMean_myl5_183
225
+ 190 kgen Float __myl_AddSqrtDivMulMulSlicNegSlicConcMulMulAddReshSlicReplReshTran_myl5_184
226
+ 191 kgen Float __myl_TranReshTranMulMean_myl5_185
227
+ 192 kgen Float __myl_AddSqrtDivMulMulSlicNegSlicConcMulMulAdd_myl5_186
228
+ 193 gemm Float /text_model/layers_8/self_attn/MatMul_myl5_187
229
+ 194 kgen Float __myl_SlicReshAddMaxrSubExpSumDivMul_myl5_188
230
+ 195 kgen Float __myl_MoveReplConc_myl5_189
231
+ 196 kgen Float __myl_MoveReplConc_myl5_190
232
+ 197 gemm Float /text_model/layers_8/self_attn/MatMul_1_myl5_191
233
+ 198 kgen FP8 __myl_MulCast_myl5_192
234
+ 199 gemm Float /text_model/layers_8/self_attn/o_proj/MatMul_myl5_193
235
+ 200 kgen Float __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulMulCastMulCast_myl5_194
236
+ 201 gemm Float /text_model/layers_8/mlp/up_proj/MatMul_myl5_195
237
+ 202 gemm Float /text_model/layers_8/mlp/gate_proj/MatMul_myl5_196
238
+ 203 kgen FP8 __myl_MulMulMulAddMulTanhAddMulMulMulMulCast_myl5_197
239
+ 204 gemm Float /text_model/layers_8/mlp/down_proj/MatMul_myl5_198
240
+ 205 kgen Float __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulMulCastMulCastMulCast_myl5_199
241
+ 206 gemm Float /text_model/layers_9/self_attn/v_proj/MatMul_myl5_200
242
+ 207 kgen Float __myl_TranReshReshSlicRepl_myl5_201
243
+ 208 gemm Float /text_model/layers_9/self_attn/k_proj/MatMul_myl5_202
244
+ 209 gemm Float /text_model/layers_9/self_attn/q_proj/MatMul_myl5_203
245
+ 210 kgen Float __myl_TranReshReshMulMean_myl5_204
246
+ 211 kgen Float __myl_AddSqrtDivMulMulSlicNegSlicConcMulMulAddReshSlicReplReshTran_myl5_205
247
+ 212 kgen Float __myl_TranReshTranMulMean_myl5_206
248
+ 213 kgen Float __myl_AddSqrtDivMulMulSlicNegSlicConcMulMulAdd_myl5_207
249
+ 214 gemm Float /text_model/layers_9/self_attn/MatMul_myl5_208
250
+ 215 kgen Float __myl_SlicReshAddMaxrSubExpSumDivMul_myl5_209
251
+ 216 kgen Float __myl_MoveReplConc_myl5_210
252
+ 217 kgen Float __myl_MoveReplConc_myl5_211
253
+ 218 gemm Float /text_model/layers_9/self_attn/MatMul_1_myl5_212
254
+ 219 kgen FP8 __myl_MulCast_myl5_213
255
+ 220 gemm Float /text_model/layers_9/self_attn/o_proj/MatMul_myl5_214
256
+ 221 kgen Float __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulMulCastMulCast_myl5_215
257
+ 222 gemm Float /text_model/layers_9/mlp/up_proj/MatMul_myl5_216
258
+ 223 gemm Float /text_model/layers_9/mlp/gate_proj/MatMul_myl5_217
259
+ 224 kgen FP8 __myl_MulMulMulAddMulTanhAddMulMulMulMulCast_myl5_218
260
+ 225 gemm Float /text_model/layers_9/mlp/down_proj/MatMul_myl5_219
261
+ 226 kgen Float __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulMulCastMulCastMulCast_myl5_220
262
+ 227 gemm Float /text_model/layers_10/self_attn/v_proj/MatMul_myl5_221
263
+ 228 kgen Float __myl_TranReshReshSlicRepl_myl5_222
264
+ 229 gemm Float /text_model/layers_10/self_attn/k_proj/MatMul_myl5_223
265
+ 230 gemm Float /text_model/layers_10/self_attn/q_proj/MatMul_myl5_224
266
+ 231 kgen Float __myl_TranReshReshMulMean_myl5_225
267
+ 232 kgen Float __myl_AddSqrtDivMulMulSlicNegSlicConcMulMulAddReshSlicReplReshTran_myl5_226
268
+ 233 kgen Float __myl_TranReshTranMulMean_myl5_227
269
+ 234 kgen Float __myl_AddSqrtDivMulMulSlicNegSlicConcMulMulAdd_myl5_228
270
+ 235 gemm Float /text_model/layers_10/self_attn/MatMul_myl5_229
271
+ 236 kgen Float __myl_SlicReshAddMaxrSubExpSumDivMul_myl5_230
272
+ 237 kgen Float __myl_MoveReplConc_myl5_231
273
+ 238 kgen Float __myl_MoveReplConc_myl5_232
274
+ 239 gemm Float /text_model/layers_10/self_attn/MatMul_1_myl5_233
275
+ 240 kgen FP8 __myl_MulCast_myl5_234
276
+ 241 gemm Float /text_model/layers_10/self_attn/o_proj/MatMul_myl5_235
277
+ 242 kgen Float __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulMulCastMulCast_myl5_236
278
+ 243 gemm Float /text_model/layers_10/mlp/up_proj/MatMul_myl5_237
279
+ 244 gemm Float /text_model/layers_10/mlp/gate_proj/MatMul_myl5_238
280
+ 245 kgen FP8 __myl_MulMulMulAddMulTanhAddMulMulMulMulCast_myl5_239
281
+ 246 gemm Float /text_model/layers_10/mlp/down_proj/MatMul_myl5_240
282
+ 247 kgen Float __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulMulCastMulCastMulCast_myl5_241
283
+ 248 gemm Float /text_model/layers_11/self_attn/v_proj/MatMul_myl5_242
284
+ 249 kgen Float __myl_TranReshReshSlicRepl_myl5_243
285
+ 250 gemm Float /text_model/layers_11/self_attn/k_proj/MatMul_myl5_244
286
+ 251 gemm Float /text_model/layers_11/self_attn/q_proj/MatMul_myl5_245
287
+ 252 kgen Float __myl_TranReshReshMulMean_myl5_246
288
+ 253 kgen Float __myl_AddSqrtDivMulMulSlicNegSlicConcMulMulAddReshSlicReplReshTran_myl5_247
289
+ 254 kgen Float __myl_TranReshTranMulMean_myl5_248
290
+ 255 kgen Float __myl_AddSqrtDivMulMulSlicNegSlicConcMulMulAdd_myl5_249
291
+ 256 gemm Float /text_model/layers_11/self_attn/MatMul_myl5_250
292
+ 257 kgen Float __myl_SlicReshAddMaxrSubExpSumDivMul_myl5_251
293
+ 258 kgen Float __myl_MoveReplConc_myl5_252
294
+ 259 kgen Float __myl_MoveReplConc_myl5_253
295
+ 260 gemm Float /text_model/layers_11/self_attn/MatMul_1_myl5_254
296
+ 261 kgen FP8 __myl_MulCast_myl5_255
297
+ 262 gemm Float /text_model/layers_11/self_attn/o_proj/MatMul_myl5_256
298
+ 263 kgen Float __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulMulCastMulCast_myl5_257
299
+ 264 gemm Float /text_model/layers_11/mlp/up_proj/MatMul_myl5_258
300
+ 265 gemm Float /text_model/layers_11/mlp/gate_proj/MatMul_myl5_259
301
+ 266 kgen FP8 __myl_MulMulMulAddMulTanhAddMulMulMulMulCast_myl5_260
302
+ 267 gemm Float /text_model/layers_11/mlp/down_proj/MatMul_myl5_261
303
+ 268 kgen Float __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulMulCastMulCastMulCast_myl5_262
304
+ 269 gemm Float /text_model/layers_12/self_attn/v_proj/MatMul_myl5_263
305
+ 270 kgen Float __myl_TranReshReshSlicRepl_myl5_264
306
+ 271 gemm Float /text_model/layers_12/self_attn/k_proj/MatMul_myl5_265
307
+ 272 gemm Float /text_model/layers_12/self_attn/q_proj/MatMul_myl5_266
308
+ 273 kgen Float __myl_TranReshReshMulMean_myl5_267
309
+ 274 kgen Float __myl_AddSqrtDivMulMulSlicNegSlicConcMulMulAddReshSlicReplReshTran_myl5_268
310
+ 275 kgen Float __myl_TranReshTranMulMean_myl5_269
311
+ 276 kgen Float __myl_AddSqrtDivMulMulSlicNegSlicConcMulMulAdd_myl5_270
312
+ 277 gemm Float /text_model/layers_12/self_attn/MatMul_myl5_271
313
+ 278 kgen Float __myl_SlicReshAddMaxrSubExpSumDivMul_myl5_272
314
+ 279 kgen Float __myl_MoveReplConc_myl5_273
315
+ 280 kgen Float __myl_MoveReplConc_myl5_274
316
+ 281 gemm Float /text_model/layers_12/self_attn/MatMul_1_myl5_275
317
+ 282 kgen FP8 __myl_MulCast_myl5_276
318
+ 283 gemm Float /text_model/layers_12/self_attn/o_proj/MatMul_myl5_277
319
+ 284 kgen Float __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulMulCastMulCast_myl5_278
320
+ 285 gemm Float /text_model/layers_12/mlp/up_proj/MatMul_myl5_279
321
+ 286 gemm Float /text_model/layers_12/mlp/gate_proj/MatMul_myl5_280
322
+ 287 kgen FP8 __myl_MulMulMulAddMulTanhAddMulMulMulMulCast_myl5_281
323
+ 288 gemm Float /text_model/layers_12/mlp/down_proj/MatMul_myl5_282
324
+ 289 kgen Float __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulMulCastMulCastMulCast_myl5_283
325
+ 290 gemm Float /text_model/layers_13/self_attn/v_proj/MatMul_myl5_284
326
+ 291 kgen Float __myl_TranReshReshSlicRepl_myl5_285
327
+ 292 gemm Float /text_model/layers_13/self_attn/k_proj/MatMul_myl5_286
328
+ 293 gemm Float /text_model/layers_13/self_attn/q_proj/MatMul_myl5_287
329
+ 294 kgen Float __myl_TranReshReshMulMean_myl5_288
330
+ 295 kgen Float __myl_AddSqrtDivMulMulSlicNegSlicConcMulMulAddReshSlicReplReshTran_myl5_289
331
+ 296 kgen Float __myl_TranReshTranMulMean_myl5_290
332
+ 297 kgen Float __myl_AddSqrtDivMulMulSlicNegSlicConcMulMulAdd_myl5_291
333
+ 298 gemm Float /text_model/layers_13/self_attn/MatMul_myl5_292
334
+ 299 kgen Float __myl_SlicReshAddMaxrSubExpSumDivMul_myl5_293
335
+ 300 kgen Float __myl_MoveReplConc_myl5_294
336
+ 301 kgen Float __myl_MoveReplConc_myl5_295
337
+ 302 gemm Float /text_model/layers_13/self_attn/MatMul_1_myl5_296
338
+ 303 kgen FP8 __myl_MulCast_myl5_297
339
+ 304 gemm Float /text_model/layers_13/self_attn/o_proj/MatMul_myl5_298
340
+ 305 kgen Float __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulMulCastMulCast_myl5_299
341
+ 306 gemm Float /text_model/layers_13/mlp/up_proj/MatMul_myl5_300
342
+ 307 gemm Float /text_model/layers_13/mlp/gate_proj/MatMul_myl5_301
343
+ 308 kgen FP8 __myl_MulMulMulAddMulTanhAddMulMulMulMulCast_myl5_302
344
+ 309 gemm Float /text_model/layers_13/mlp/down_proj/MatMul_myl5_303
345
+ 310 kgen Float __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulMulCastMulCastMulCast_myl5_304
346
+ 311 gemm Float /text_model/layers_14/self_attn/v_proj/MatMul_myl5_305
347
+ 312 kgen Float __myl_TranReshReshSlicRepl_myl5_306
348
+ 313 gemm Float /text_model/layers_14/self_attn/k_proj/MatMul_myl5_307
349
+ 314 gemm Float /text_model/layers_14/self_attn/q_proj/MatMul_myl5_308
350
+ 315 kgen Float __myl_TranReshReshMulMean_myl5_309
351
+ 316 kgen Float __myl_AddSqrtDivMulMulSlicNegSlicConcMulMulAddReshSlicReplReshTran_myl5_310
352
+ 317 kgen Float __myl_TranReshTranMulMean_myl5_311
353
+ 318 kgen Float __myl_AddSqrtDivMulMulSlicNegSlicConcMulMulAdd_myl5_312
354
+ 319 gemm Float /text_model/layers_14/self_attn/MatMul_myl5_313
355
+ 320 kgen Float __myl_SlicReshAddMaxrSubExpSumDivMul_myl5_314
356
+ 321 kgen Float __myl_MoveReplConc_myl5_315
357
+ 322 kgen Float __myl_MoveReplConc_myl5_316
358
+ 323 gemm Float /text_model/layers_14/self_attn/MatMul_1_myl5_317
359
+ 324 kgen FP8 __myl_MulCast_myl5_318
360
+ 325 gemm Float /text_model/layers_14/self_attn/o_proj/MatMul_myl5_319
361
+ 326 kgen Float __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulMulCastMulCast_myl5_320
362
+ 327 gemm Float /text_model/layers_14/mlp/up_proj/MatMul_myl5_321
363
+ 328 gemm Float /text_model/layers_14/mlp/gate_proj/MatMul_myl5_322
364
+ 329 kgen FP8 __myl_MulMulMulAddMulTanhAddMulMulMulMulCast_myl5_323
365
+ 330 gemm Float /text_model/layers_14/mlp/down_proj/MatMul_myl5_324
366
+ 331 kgen Float __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulMulCastMulCastMulCast_myl5_325
367
+ 332 gemm Float /text_model/layers_15/self_attn/v_proj/MatMul_myl5_326
368
+ 333 kgen Float __myl_TranReshReshSlicRepl_myl5_327
369
+ 334 gemm Float /text_model/layers_15/self_attn/k_proj/MatMul_myl5_328
370
+ 335 gemm Float /text_model/layers_15/self_attn/q_proj/MatMul_myl5_329
371
+ 336 kgen Float __myl_TranReshReshMulMean_myl5_330
372
+ 337 kgen Float __myl_AddSqrtDivMulMulSlicNegSlicConcMulMulAddReshSlicReplReshTran_myl5_331
373
+ 338 kgen Float __myl_TranReshTranMulMean_myl5_332
374
+ 339 kgen Float __myl_AddSqrtDivMulMulSlicNegSlicConcMulMulAdd_myl5_333
375
+ 340 gemm Float /text_model/layers_15/self_attn/MatMul_myl5_334
376
+ 341 kgen Float __myl_SlicReshAddMaxrSubExpSumDivMul_myl5_335
377
+ 342 kgen Float __myl_MoveReplConc_myl5_336
378
+ 343 kgen Float __myl_MoveReplConc_myl5_337
379
+ 344 gemm Float /text_model/layers_15/self_attn/MatMul_1_myl5_338
380
+ 345 kgen FP8 __myl_MulCast_myl5_339
381
+ 346 gemm Float /text_model/layers_15/self_attn/o_proj/MatMul_myl5_340
382
+ 347 kgen Float __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulMulCastMulCast_myl5_341
383
+ 348 gemm Float /text_model/layers_15/mlp/up_proj/MatMul_myl5_342
384
+ 349 gemm Float /text_model/layers_15/mlp/gate_proj/MatMul_myl5_343
385
+ 350 kgen FP8 __myl_MulMulMulAddMulTanhAddMulMulMulMulCast_myl5_344
386
+ 351 gemm Float /text_model/layers_15/mlp/down_proj/MatMul_myl5_345
387
+ 352 kgen Float __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulMulCastMulCastMulCast_myl5_346
388
+ 353 gemm Float /text_model/layers_16/self_attn/v_proj/MatMul_myl5_347
389
+ 354 kgen Float __myl_TranReshReshSlicRepl_myl5_348
390
+ 355 gemm Float /text_model/layers_16/self_attn/k_proj/MatMul_myl5_349
391
+ 356 gemm Float /text_model/layers_16/self_attn/q_proj/MatMul_myl5_350
392
+ 357 kgen Float __myl_TranReshReshMulMean_myl5_351
393
+ 358 kgen Float __myl_AddSqrtDivMulMulSlicNegSlicConcMulMulAddReshSlicReplReshTran_myl5_352
394
+ 359 kgen Float __myl_TranReshTranMulMean_myl5_353
395
+ 360 kgen Float __myl_AddSqrtDivMulMulSlicNegSlicConcMulMulAdd_myl5_354
396
+ 361 gemm Float /text_model/layers_16/self_attn/MatMul_myl5_355
397
+ 362 kgen Float __myl_SlicReshAddMaxrSubExpSumDivMul_myl5_356
398
+ 363 kgen Float __myl_MoveReplConc_myl5_357
399
+ 364 kgen Float __myl_MoveReplConc_myl5_358
400
+ 365 gemm Float /text_model/layers_16/self_attn/MatMul_1_myl5_359
401
+ 366 kgen FP8 __myl_MulCast_myl5_360
402
+ 367 gemm Float /text_model/layers_16/self_attn/o_proj/MatMul_myl5_361
403
+ 368 kgen Float __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulMulCastMulCast_myl5_362
404
+ 369 gemm Float /text_model/layers_16/mlp/up_proj/MatMul_myl5_363
405
+ 370 gemm Float /text_model/layers_16/mlp/gate_proj/MatMul_myl5_364
406
+ 371 kgen FP8 __myl_MulMulMulAddMulTanhAddMulMulMulMulCast_myl5_365
407
+ 372 gemm Float /text_model/layers_16/mlp/down_proj/MatMul_myl5_366
408
+ 373 kgen Float __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulMulCastMulCastMulCast_myl5_367
409
+ 374 gemm Float /text_model/layers_17/self_attn/v_proj/MatMul_myl5_368
410
+ 375 kgen Float __myl_TranReshReshSlicRepl_myl5_369
411
+ 376 gemm Float /text_model/layers_17/self_attn/k_proj/MatMul_myl5_370
412
+ 377 gemm Float /text_model/layers_17/self_attn/q_proj/MatMul_myl5_371
413
+ 378 kgen Float __myl_TranReshReshMulMean_myl5_372
414
+ 379 kgen Float __myl_AddSqrtDivMulMulSlicNegSlicConcMulMulAddReshSlicReplReshTran_myl5_373
415
+ 380 kgen Float __myl_TranReshTranMulMean_myl5_374
416
+ 381 kgen Float __myl_AddSqrtDivMulMulSlicNegSlicConcMulMulAdd_myl5_375
417
+ 382 gemm Float /text_model/layers_17/self_attn/MatMul_myl5_376
418
+ 383 kgen Float __myl_SlicReshAddMaxrSubExpSumDivMul_myl5_377
419
+ 384 kgen Float __myl_MoveReplConc_myl5_378
420
+ 385 kgen Float __myl_MoveReplConc_myl5_379
421
+ 386 gemm Float /text_model/layers_17/self_attn/MatMul_1_myl5_380
422
+ 387 kgen FP8 __myl_MulCast_myl5_381
423
+ 388 gemm Float /text_model/layers_17/self_attn/o_proj/MatMul_myl5_382
424
+ 389 kgen Float __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulMulCastMulCast_myl5_383
425
+ 390 gemm Float /text_model/layers_17/mlp/up_proj/MatMul_myl5_384
426
+ 391 gemm Float /text_model/layers_17/mlp/gate_proj/MatMul_myl5_385
427
+ 392 kgen FP8 __myl_MulMulMulAddMulTanhAddMulMulMulMulCast_myl5_386
428
+ 393 gemm Float /text_model/layers_17/mlp/down_proj/MatMul_myl5_387
429
+ 394 kgen Float __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulMulCastMulCastMulCast_myl5_388
430
+ 395 gemm Float /text_model/layers_18/self_attn/v_proj/MatMul_myl5_389
431
+ 396 kgen Float __myl_TranReshReshSlicRepl_myl5_390
432
+ 397 gemm Float /text_model/layers_18/self_attn/k_proj/MatMul_myl5_391
433
+ 398 gemm Float /text_model/layers_18/self_attn/q_proj/MatMul_myl5_392
434
+ 399 kgen Float __myl_TranReshReshMulMean_myl5_393
435
+ 400 kgen Float __myl_AddSqrtDivMulMulSlicNegSlicConcMulMulAddReshSlicReplReshTran_myl5_394
436
+ 401 kgen Float __myl_TranReshTranMulMean_myl5_395
437
+ 402 kgen Float __myl_AddSqrtDivMulMulSlicNegSlicConcMulMulAdd_myl5_396
438
+ 403 gemm Float /text_model/layers_18/self_attn/MatMul_myl5_397
439
+ 404 kgen Float __myl_SlicReshAddMaxrSubExpSumDivMul_myl5_398
440
+ 405 kgen Float __myl_MoveReplConc_myl5_399
441
+ 406 kgen Float __myl_MoveReplConc_myl5_400
442
+ 407 gemm Float /text_model/layers_18/self_attn/MatMul_1_myl5_401
443
+ 408 kgen FP8 __myl_MulCast_myl5_402
444
+ 409 gemm Float /text_model/layers_18/self_attn/o_proj/MatMul_myl5_403
445
+ 410 kgen Float __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulMulCastMulCast_myl5_404
446
+ 411 gemm Float /text_model/layers_18/mlp/up_proj/MatMul_myl5_405
447
+ 412 gemm Float /text_model/layers_18/mlp/gate_proj/MatMul_myl5_406
448
+ 413 kgen FP8 __myl_MulMulMulAddMulTanhAddMulMulMulMulCast_myl5_407
449
+ 414 gemm Float /text_model/layers_18/mlp/down_proj/MatMul_myl5_408
450
+ 415 kgen Float __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulMulCastMulCastMulCast_myl5_409
451
+ 416 gemm Float /text_model/layers_19/self_attn/v_proj/MatMul_myl5_410
452
+ 417 kgen Float __myl_TranReshReshSlicRepl_myl5_411
453
+ 418 gemm Float /text_model/layers_19/self_attn/k_proj/MatMul_myl5_412
454
+ 419 gemm Float /text_model/layers_19/self_attn/q_proj/MatMul_myl5_413
455
+ 420 kgen Float __myl_TranReshReshMulMean_myl5_414
456
+ 421 kgen Float __myl_AddSqrtDivMulMulSlicNegSlicConcMulMulAddReshSlicReplReshTran_myl5_415
457
+ 422 kgen Float __myl_TranReshTranMulMean_myl5_416
458
+ 423 kgen Float __myl_AddSqrtDivMulMulSlicNegSlicConcMulMulAdd_myl5_417
459
+ 424 gemm Float /text_model/layers_19/self_attn/MatMul_myl5_418
460
+ 425 kgen Float __myl_SlicReshAddMaxrSubExpSumDivMul_myl5_419
461
+ 426 kgen Float __myl_MoveReplConc_myl5_420
462
+ 427 kgen Float __myl_MoveReplConc_myl5_421
463
+ 428 gemm Float /text_model/layers_19/self_attn/MatMul_1_myl5_422
464
+ 429 kgen FP8 __myl_MulCast_myl5_423
465
+ 430 gemm Float /text_model/layers_19/self_attn/o_proj/MatMul_myl5_424
466
+ 431 kgen Float __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulMulCastMulCast_myl5_425
467
+ 432 gemm Float /text_model/layers_19/mlp/up_proj/MatMul_myl5_426
468
+ 433 gemm Float /text_model/layers_19/mlp/gate_proj/MatMul_myl5_427
469
+ 434 kgen FP8 __myl_MulMulMulAddMulTanhAddMulMulMulMulCast_myl5_428
470
+ 435 gemm Float /text_model/layers_19/mlp/down_proj/MatMul_myl5_429
471
+ 436 kgen Float __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulMulCastMulCastMulCast_myl5_430
472
+ 437 gemm Float /text_model/layers_20/self_attn/v_proj/MatMul_myl5_431
473
+ 438 kgen Float __myl_TranReshReshSlicRepl_myl5_432
474
+ 439 gemm Float /text_model/layers_20/self_attn/k_proj/MatMul_myl5_433
475
+ 440 gemm Float /text_model/layers_20/self_attn/q_proj/MatMul_myl5_434
476
+ 441 kgen Float __myl_TranReshReshMulMean_myl5_435
477
+ 442 kgen Float __myl_AddSqrtDivMulMulSlicNegSlicConcMulMulAddReshSlicReplReshTran_myl5_436
478
+ 443 kgen Float __myl_TranReshTranMulMean_myl5_437
479
+ 444 kgen Float __myl_AddSqrtDivMulMulSlicNegSlicConcMulMulAdd_myl5_438
480
+ 445 gemm Float /text_model/layers_20/self_attn/MatMul_myl5_439
481
+ 446 kgen Float __myl_SlicReshAddMaxrSubExpSumDivMul_myl5_440
482
+ 447 kgen Float __myl_MoveReplConc_myl5_441
483
+ 448 kgen Float __myl_MoveReplConc_myl5_442
484
+ 449 gemm Float /text_model/layers_20/self_attn/MatMul_1_myl5_443
485
+ 450 kgen FP8 __myl_MulCast_myl5_444
486
+ 451 gemm Float /text_model/layers_20/self_attn/o_proj/MatMul_myl5_445
487
+ 452 kgen Float __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulMulCastMulCast_myl5_446
488
+ 453 gemm Float /text_model/layers_20/mlp/up_proj/MatMul_myl5_447
489
+ 454 gemm Float /text_model/layers_20/mlp/gate_proj/MatMul_myl5_448
490
+ 455 kgen FP8 __myl_MulMulMulAddMulTanhAddMulMulMulMulCast_myl5_449
491
+ 456 gemm Float /text_model/layers_20/mlp/down_proj/MatMul_myl5_450
492
+ 457 kgen Float __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulMulCastMulCastMulCast_myl5_451
493
+ 458 gemm Float /text_model/layers_21/self_attn/v_proj/MatMul_myl5_452
494
+ 459 kgen Float __myl_TranReshReshSlicRepl_myl5_453
495
+ 460 gemm Float /text_model/layers_21/self_attn/k_proj/MatMul_myl5_454
496
+ 461 gemm Float /text_model/layers_21/self_attn/q_proj/MatMul_myl5_455
497
+ 462 kgen Float __myl_TranReshReshMulMean_myl5_456
498
+ 463 kgen Float __myl_AddSqrtDivMulMulSlicNegSlicConcMulMulAddReshSlicReplReshTran_myl5_457
499
+ 464 kgen Float __myl_TranReshTranMulMean_myl5_458
500
+ 465 kgen Float __myl_AddSqrtDivMulMulSlicNegSlicConcMulMulAdd_myl5_459
501
+ 466 gemm Float /text_model/layers_21/self_attn/MatMul_myl5_460
502
+ 467 kgen Float __myl_SlicReshAddMaxrSubExpSumDivMul_myl5_461
503
+ 468 kgen Float __myl_MoveReplConc_myl5_462
504
+ 469 kgen Float __myl_MoveReplConc_myl5_463
505
+ 470 gemm Float /text_model/layers_21/self_attn/MatMul_1_myl5_464
506
+ 471 kgen FP8 __myl_MulCast_myl5_465
507
+ 472 gemm Float /text_model/layers_21/self_attn/o_proj/MatMul_myl5_466
508
+ 473 kgen Float __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulMulCastMulCast_myl5_467
509
+ 474 gemm Float /text_model/layers_21/mlp/up_proj/MatMul_myl5_468
510
+ 475 gemm Float /text_model/layers_21/mlp/gate_proj/MatMul_myl5_469
511
+ 476 kgen FP8 __myl_MulMulMulAddMulTanhAddMulMulMulMulCast_myl5_470
512
+ 477 gemm Float /text_model/layers_21/mlp/down_proj/MatMul_myl5_471
513
+ 478 kgen Float __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulMulCastMulCastMulCast_myl5_472
514
+ 479 gemm Float /text_model/layers_22/self_attn/v_proj/MatMul_myl5_473
515
+ 480 kgen Float __myl_TranReshReshSlicRepl_myl5_474
516
+ 481 gemm Float /text_model/layers_22/self_attn/k_proj/MatMul_myl5_475
517
+ 482 gemm Float /text_model/layers_22/self_attn/q_proj/MatMul_myl5_476
518
+ 483 kgen Float __myl_TranReshReshMulMean_myl5_477
519
+ 484 kgen Float __myl_AddSqrtDivMulMulSlicNegSlicConcMulMulAddReshSlicReplReshTran_myl5_478
520
+ 485 kgen Float __myl_TranReshTranMulMean_myl5_479
521
+ 486 kgen Float __myl_AddSqrtDivMulMulSlicNegSlicConcMulMulAdd_myl5_480
522
+ 487 gemm Float /text_model/layers_22/self_attn/MatMul_myl5_481
523
+ 488 kgen Float __myl_SlicReshAddMaxrSubExpSumDivMul_myl5_482
524
+ 489 kgen Float __myl_MoveReplConc_myl5_483
525
+ 490 kgen Float __myl_MoveReplConc_myl5_484
526
+ 491 gemm Float /text_model/layers_22/self_attn/MatMul_1_myl5_485
527
+ 492 kgen FP8 __myl_MulCast_myl5_486
528
+ 493 gemm Float /text_model/layers_22/self_attn/o_proj/MatMul_myl5_487
529
+ 494 kgen Float __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulMulCastMulCast_myl5_488
530
+ 495 gemm Float /text_model/layers_22/mlp/up_proj/MatMul_myl5_489
531
+ 496 gemm Float /text_model/layers_22/mlp/gate_proj/MatMul_myl5_490
532
+ 497 kgen FP8 __myl_MulMulMulAddMulTanhAddMulMulMulMulCast_myl5_491
533
+ 498 gemm Float /text_model/layers_22/mlp/down_proj/MatMul_myl5_492
534
+ 499 kgen Float __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulMulCastMulCastMulCast_myl5_493
535
+ 500 gemm Float /text_model/layers_23/self_attn/v_proj/MatMul_myl5_494
536
+ 501 kgen Float __myl_TranReshReshSlicRepl_myl5_495
537
+ 502 gemm Float /text_model/layers_23/self_attn/k_proj/MatMul_myl5_496
538
+ 503 gemm Float /text_model/layers_23/self_attn/q_proj/MatMul_myl5_497
539
+ 504 kgen Float __myl_TranReshReshMulMean_myl5_498
540
+ 505 kgen Float __myl_AddSqrtDivMulMulSlicNegSlicConcMulMulAddReshSlicReplReshTran_myl5_499
541
+ 506 kgen Float __myl_TranReshTranMulMean_myl5_500
542
+ 507 kgen Float __myl_AddSqrtDivMulMulSlicNegSlicConcMulMulAdd_myl5_501
543
+ 508 gemm Float /text_model/layers_23/self_attn/MatMul_myl5_502
544
+ 509 kgen Float __myl_SlicReshAddMaxrSubExpSumDivMul_myl5_503
545
+ 510 kgen Float __myl_MoveReplConc_myl5_504
546
+ 511 kgen Float __myl_MoveReplConc_myl5_505
547
+ 512 gemm Float /text_model/layers_23/self_attn/MatMul_1_myl5_506
548
+ 513 kgen FP8 __myl_MulCast_myl5_507
549
+ 514 gemm Float /text_model/layers_23/self_attn/o_proj/MatMul_myl5_508
550
+ 515 kgen Float __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulMulCastMulCast_myl5_509
551
+ 516 gemm Float /text_model/layers_23/mlp/up_proj/MatMul_myl5_510
552
+ 517 gemm Float /text_model/layers_23/mlp/gate_proj/MatMul_myl5_511
553
+ 518 kgen FP8 __myl_MulMulMulAddMulTanhAddMulMulMulMulCast_myl5_512
554
+ 519 gemm Float /text_model/layers_23/mlp/down_proj/MatMul_myl5_513
555
+ 520 kgen Float __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulMulCastMulCastMulCast_myl5_514
556
+ 521 gemm Float /text_model/layers_24/self_attn/v_proj/MatMul_myl5_515
557
+ 522 kgen Float __myl_TranReshReshSlicRepl_myl5_516
558
+ 523 gemm Float /text_model/layers_24/self_attn/k_proj/MatMul_myl5_517
559
+ 524 gemm Float /text_model/layers_24/self_attn/q_proj/MatMul_myl5_518
560
+ 525 kgen Float __myl_TranReshReshMulMean_myl5_519
561
+ 526 kgen Float __myl_AddSqrtDivMulMulSlicNegSlicConcMulMulAddReshSlicReplReshTran_myl5_520
562
+ 527 kgen Float __myl_TranReshTranMulMean_myl5_521
563
+ 528 kgen Float __myl_AddSqrtDivMulMulSlicNegSlicConcMulMulAdd_myl5_522
564
+ 529 gemm Float /text_model/layers_24/self_attn/MatMul_myl5_523
565
+ 530 kgen Float __myl_SlicReshAddMaxrSubExpSumDivMul_myl5_524
566
+ 531 kgen Float __myl_MoveReplConc_myl5_525
567
+ 532 kgen Float __myl_MoveReplConc_myl5_526
568
+ 533 gemm Float /text_model/layers_24/self_attn/MatMul_1_myl5_527
569
+ 534 kgen FP8 __myl_MulCast_myl5_528
570
+ 535 gemm Float /text_model/layers_24/self_attn/o_proj/MatMul_myl5_529
571
+ 536 kgen Float __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulMulCastMulCast_myl5_530
572
+ 537 gemm Float /text_model/layers_24/mlp/up_proj/MatMul_myl5_531
573
+ 538 gemm Float /text_model/layers_24/mlp/gate_proj/MatMul_myl5_532
574
+ 539 kgen FP8 __myl_MulMulMulAddMulTanhAddMulMulMulMulCast_myl5_533
575
+ 540 gemm Float /text_model/layers_24/mlp/down_proj/MatMul_myl5_534
576
+ 541 kgen Float __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulMulCastMulCastMulCast_myl5_535
577
+ 542 gemm Float /text_model/layers_25/self_attn/v_proj/MatMul_myl5_536
578
+ 543 kgen Float __myl_TranReshReshSlicRepl_myl5_537
579
+ 544 gemm Float /text_model/layers_25/self_attn/k_proj/MatMul_myl5_538
580
+ 545 gemm Float /text_model/layers_25/self_attn/q_proj/MatMul_myl5_539
581
+ 546 kgen Float __myl_TranReshReshMulMean_myl5_540
582
+ 547 kgen Float __myl_AddSqrtDivMulMulSlicNegSlicConcMulMulAddReshSlicReplReshTran_myl5_541
583
+ 548 kgen Float __myl_TranReshTranMulMean_myl5_542
584
+ 549 kgen Float __myl_AddSqrtDivMulMulSlicNegSlicConcMulMulAdd_myl5_543
585
+ 550 gemm Float /text_model/layers_25/self_attn/MatMul_myl5_544
586
+ 551 kgen Float __myl_SlicReshAddMaxrSubExpSumDivMul_myl5_545
587
+ 552 kgen Float __myl_MoveReplConc_myl5_546
588
+ 553 kgen Float __myl_MoveReplConc_myl5_547
589
+ 554 gemm Float /text_model/layers_25/self_attn/MatMul_1_myl5_548
590
+ 555 kgen FP8 __myl_MulCast_myl5_549
591
+ 556 kgen Float __myl_CastSum_myl5_550
592
+ 557 gemm Float /text_model/layers_25/self_attn/o_proj/MatMul_myl5_551
593
+ 558 kgen Float __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulMulCastMulCast_myl5_552
594
+ 559 gemm Float /text_model/layers_25/mlp/up_proj/MatMul_myl5_553
595
+ 560 gemm Float /text_model/layers_25/mlp/gate_proj/MatMul_myl5_554
596
+ 561 kgen FP8 __myl_MulMulMulAddMulTanhAddMulMulMulMulCast_myl5_555
597
+ 562 gemm Float /text_model/layers_25/mlp/down_proj/MatMul_myl5_556
598
+ 563 kgen Float __myl_MulMeanAddSqrtDivMulMulAddMulMean_myl5_557
599
+ 564 kgen FP8 __myl_ReshMaxMinAddSqrtDivMulMulMoveReshTranReshMoveReshMulSumReshDivMulMulCast_myl5_558
600
+ 565 gemm Float /classifier/Gemm_myl5_559
601
+ 566 kgen Float __myl_Slic_myl5_560
602
+ 567 TrainStation - [trainStation3]