Upload TRT model for Nvidia L40S (FP8)
#14
by sarah-cisco - opened
- .gitattributes +1 -0
- model_l40s_fp8.plan +3 -0
- trt_engine_layer_summary_l40s_fp8.txt +501 -0
.gitattributes
CHANGED
|
@@ -40,3 +40,4 @@ model_h200_bf16.plan filter=lfs diff=lfs merge=lfs -text
|
|
| 40 |
model_h100_bf16.plan filter=lfs diff=lfs merge=lfs -text
|
| 41 |
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
|
|
|
|
|
|
| 40 |
model_h100_bf16.plan filter=lfs diff=lfs merge=lfs -text
|
| 41 |
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
|
model_l40s_fp8.plan
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:590ea068d27c09ff8da676dcb5129b271e748d2e43ac4975558eebc1c1ee3c4b
|
| 3 |
+
size 1312109860
|
trt_engine_layer_summary_l40s_fp8.txt
ADDED
|
@@ -0,0 +1,501 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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: 466
|
| 14 |
+
|
| 15 |
+
==== Precision Statistics ====
|
| 16 |
+
BFloat16: 1193
|
| 17 |
+
FP8: 340
|
| 18 |
+
Int64: 14
|
| 19 |
+
Bool: 2
|
| 20 |
+
Int32: 2
|
| 21 |
+
Float: 1
|
| 22 |
+
|
| 23 |
+
==== Layer Type Statistics ====
|
| 24 |
+
kgen: 271
|
| 25 |
+
gemm: 183
|
| 26 |
+
TrainStation: 3
|
| 27 |
+
shape_call: 3
|
| 28 |
+
reshape: 2
|
| 29 |
+
cast: 2
|
| 30 |
+
NonZero: 1
|
| 31 |
+
DeviceToShapeHost: 1
|
| 32 |
+
|
| 33 |
+
==== Layer List ====
|
| 34 |
+
# Type Precision Name
|
| 35 |
+
----------------------------------------------------------------------------------------------------
|
| 36 |
+
0 TrainStation - [trainStation1]
|
| 37 |
+
1 shape_call - __mye505_0_myl1_0
|
| 38 |
+
2 kgen Bool __myl_EqlReshSlicRepl_myl1_1
|
| 39 |
+
3 NonZero Int64 /text_model/embed_tokens/NonZero
|
| 40 |
+
4 DeviceToShapeHost - /text_model/embed_tokens/NonZero[size][DevicetoShapeHostCopy]
|
| 41 |
+
5 TrainStation - [trainStation2]
|
| 42 |
+
6 shape_call - __mye154348_0_myl5_0
|
| 43 |
+
7 kgen BFloat16 __myl_CastGathMul_myl5_1
|
| 44 |
+
8 kgen Int32 __myl_TranCast_myl5_2
|
| 45 |
+
9 kgen BFloat16 __myl_SlicScat_myl5_3
|
| 46 |
+
10 kgen FP8 __myl_MulMeanAddSqrtDivMulCastMulCast_myl5_4
|
| 47 |
+
11 gemm BFloat16 /text_model/layers.0/self_attn/q_proj/MatMul_myl5_5
|
| 48 |
+
12 kgen BFloat16 __myl_TranMulMean_myl5_6
|
| 49 |
+
13 gemm BFloat16 /text_model/layers.0/self_attn/k_proj/MatMul_myl5_7
|
| 50 |
+
14 kgen BFloat16 __myl_MulMeanAddSqrt_myl5_8
|
| 51 |
+
15 kgen BFloat16 __myl_IotaCastReshCast_myl5_9
|
| 52 |
+
16 kgen BFloat16 __myl_MoveMul_myl5_10
|
| 53 |
+
17 kgen BFloat16 __myl_ConcSinCosMoveReshMoveReshAddSqrtDivMulDivMulMulSlicNegSlicConcMulMulAddReshSlicReplReshEtc_myl5_11
|
| 54 |
+
18 gemm BFloat16 /text_model/layers.0/self_attn/v_proj/MatMul_myl5_12
|
| 55 |
+
19 kgen BFloat16 __myl_SlicReplReshMove_myl5_13
|
| 56 |
+
20 kgen BFloat16 __myl_ReshCastSubMulSlic_myl5_14
|
| 57 |
+
21 kgen FP8 _gemm_mha_v2_myl5_15
|
| 58 |
+
22 kgen FP8 __myl_MoveReshTran_myl5_16
|
| 59 |
+
23 gemm BFloat16 /text_model/layers.0/self_attn/o_proj/MatMul_myl5_17
|
| 60 |
+
24 kgen BFloat16 __myl_MulMeanAddSqrt_myl5_18
|
| 61 |
+
25 kgen BFloat16 __myl_DivMulMulReshAddReshMulMeanAddSqrtDivMulCastMulCast_myl5_19
|
| 62 |
+
26 gemm BFloat16 /text_model/layers.0/mlp/up_proj/MatMul_myl5_20
|
| 63 |
+
27 gemm BFloat16 /text_model/layers.0/mlp/gate_proj/MatMul_myl5_21
|
| 64 |
+
28 kgen FP8 __myl_MulMulMulAddMulTanhAddMulMulMulCastMulCast_myl5_22
|
| 65 |
+
29 gemm BFloat16 /text_model/layers.0/mlp/down_proj/MatMul_myl5_23
|
| 66 |
+
30 kgen BFloat16 __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulCastMulCast_myl5_24
|
| 67 |
+
31 gemm BFloat16 /text_model/layers.1/self_attn/k_proj/MatMul_myl5_25
|
| 68 |
+
32 kgen BFloat16 __myl_MulMeanAddSqrt_myl5_26
|
| 69 |
+
33 kgen BFloat16 __myl_DivMulMulSlicNegSlicConcMulMulAddReshSlicReplReshTranReshMove_myl5_27
|
| 70 |
+
34 gemm BFloat16 /text_model/layers.1/self_attn/q_proj/MatMul_myl5_28
|
| 71 |
+
35 kgen BFloat16 __myl_TranMulMean_myl5_29
|
| 72 |
+
36 kgen BFloat16 __myl_AddSqrtDivMulMulSlicNegSlicConcMulMulAddReshMove_myl5_30
|
| 73 |
+
37 gemm BFloat16 /text_model/layers.1/self_attn/v_proj/MatMul_myl5_31
|
| 74 |
+
38 kgen BFloat16 __myl_SlicReplReshMove_myl5_32
|
| 75 |
+
39 kgen FP8 _gemm_mha_v2_myl5_33
|
| 76 |
+
40 kgen FP8 __myl_MoveReshTran_myl5_34
|
| 77 |
+
41 gemm BFloat16 /text_model/layers.1/self_attn/o_proj/MatMul_myl5_35
|
| 78 |
+
42 kgen BFloat16 __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulCastMulCast_myl5_36
|
| 79 |
+
43 gemm BFloat16 /text_model/layers.1/mlp/up_proj/MatMul_myl5_37
|
| 80 |
+
44 gemm BFloat16 /text_model/layers.1/mlp/gate_proj/MatMul_myl5_38
|
| 81 |
+
45 kgen FP8 __myl_MulMulMulAddMulTanhAddMulMulMulCastMulCast_myl5_39
|
| 82 |
+
46 gemm BFloat16 /text_model/layers.1/mlp/down_proj/MatMul_myl5_40
|
| 83 |
+
47 kgen BFloat16 __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulCastMulCast_myl5_41
|
| 84 |
+
48 gemm BFloat16 /text_model/layers.2/self_attn/k_proj/MatMul_myl5_42
|
| 85 |
+
49 kgen BFloat16 __myl_MulMeanAddSqrt_myl5_43
|
| 86 |
+
50 kgen BFloat16 __myl_DivMulMulSlicNegSlicConcMulMulAddReshSlicReplReshTranReshMove_myl5_44
|
| 87 |
+
51 gemm BFloat16 /text_model/layers.2/self_attn/q_proj/MatMul_myl5_45
|
| 88 |
+
52 kgen BFloat16 __myl_TranMulMean_myl5_46
|
| 89 |
+
53 kgen BFloat16 __myl_AddSqrtDivMulMulSlicNegSlicConcMulMulAddReshMove_myl5_47
|
| 90 |
+
54 gemm BFloat16 /text_model/layers.2/self_attn/v_proj/MatMul_myl5_48
|
| 91 |
+
55 kgen BFloat16 __myl_SlicReplReshMove_myl5_49
|
| 92 |
+
56 kgen FP8 _gemm_mha_v2_myl5_50
|
| 93 |
+
57 kgen FP8 __myl_MoveReshTran_myl5_51
|
| 94 |
+
58 gemm BFloat16 /text_model/layers.2/self_attn/o_proj/MatMul_myl5_52
|
| 95 |
+
59 kgen BFloat16 __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulCastMulCast_myl5_53
|
| 96 |
+
60 gemm BFloat16 /text_model/layers.2/mlp/up_proj/MatMul_myl5_54
|
| 97 |
+
61 gemm BFloat16 /text_model/layers.2/mlp/gate_proj/MatMul_myl5_55
|
| 98 |
+
62 kgen FP8 __myl_MulMulMulAddMulTanhAddMulMulMulCastMulCast_myl5_56
|
| 99 |
+
63 gemm BFloat16 /text_model/layers.2/mlp/down_proj/MatMul_myl5_57
|
| 100 |
+
64 kgen BFloat16 __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulCastMulCast_myl5_58
|
| 101 |
+
65 gemm BFloat16 /text_model/layers.3/self_attn/k_proj/MatMul_myl5_59
|
| 102 |
+
66 kgen BFloat16 __myl_MulMeanAddSqrt_myl5_60
|
| 103 |
+
67 kgen BFloat16 __myl_DivMulMulSlicNegSlicConcMulMulAddReshSlicReplReshTranReshMove_myl5_61
|
| 104 |
+
68 gemm BFloat16 /text_model/layers.3/self_attn/q_proj/MatMul_myl5_62
|
| 105 |
+
69 kgen BFloat16 __myl_TranMulMean_myl5_63
|
| 106 |
+
70 kgen BFloat16 __myl_AddSqrtDivMulMulSlicNegSlicConcMulMulAddReshMove_myl5_64
|
| 107 |
+
71 gemm BFloat16 /text_model/layers.3/self_attn/v_proj/MatMul_myl5_65
|
| 108 |
+
72 kgen BFloat16 __myl_SlicReplReshMove_myl5_66
|
| 109 |
+
73 kgen FP8 _gemm_mha_v2_myl5_67
|
| 110 |
+
74 kgen FP8 __myl_MoveReshTran_myl5_68
|
| 111 |
+
75 gemm BFloat16 /text_model/layers.3/self_attn/o_proj/MatMul_myl5_69
|
| 112 |
+
76 kgen BFloat16 __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulCastMulCast_myl5_70
|
| 113 |
+
77 gemm BFloat16 /text_model/layers.3/mlp/up_proj/MatMul_myl5_71
|
| 114 |
+
78 gemm BFloat16 /text_model/layers.3/mlp/gate_proj/MatMul_myl5_72
|
| 115 |
+
79 kgen FP8 __myl_MulMulMulAddMulTanhAddMulMulMulCastMulCast_myl5_73
|
| 116 |
+
80 gemm BFloat16 /text_model/layers.3/mlp/down_proj/MatMul_myl5_74
|
| 117 |
+
81 kgen BFloat16 __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulCastMulCast_myl5_75
|
| 118 |
+
82 gemm BFloat16 /text_model/layers.4/self_attn/k_proj/MatMul_myl5_76
|
| 119 |
+
83 kgen BFloat16 __myl_MulMeanAddSqrt_myl5_77
|
| 120 |
+
84 kgen BFloat16 __myl_DivMulMulSlicNegSlicConcMulMulAddReshSlicReplReshTranReshMove_myl5_78
|
| 121 |
+
85 gemm BFloat16 /text_model/layers.4/self_attn/q_proj/MatMul_myl5_79
|
| 122 |
+
86 kgen BFloat16 __myl_TranMulMean_myl5_80
|
| 123 |
+
87 kgen BFloat16 __myl_AddSqrtDivMulMulSlicNegSlicConcMulMulAddReshMove_myl5_81
|
| 124 |
+
88 gemm BFloat16 /text_model/layers.4/self_attn/v_proj/MatMul_myl5_82
|
| 125 |
+
89 kgen BFloat16 __myl_SlicReplReshMove_myl5_83
|
| 126 |
+
90 kgen FP8 _gemm_mha_v2_myl5_84
|
| 127 |
+
91 kgen FP8 __myl_MoveReshTran_myl5_85
|
| 128 |
+
92 gemm BFloat16 /text_model/layers.4/self_attn/o_proj/MatMul_myl5_86
|
| 129 |
+
93 kgen BFloat16 __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulCastMulCast_myl5_87
|
| 130 |
+
94 gemm BFloat16 /text_model/layers.4/mlp/up_proj/MatMul_myl5_88
|
| 131 |
+
95 gemm BFloat16 /text_model/layers.4/mlp/gate_proj/MatMul_myl5_89
|
| 132 |
+
96 kgen FP8 __myl_MulMulMulAddMulTanhAddMulMulMulCastMulCast_myl5_90
|
| 133 |
+
97 gemm BFloat16 /text_model/layers.4/mlp/down_proj/MatMul_myl5_91
|
| 134 |
+
98 kgen BFloat16 __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulCastMulCast_myl5_92
|
| 135 |
+
99 gemm BFloat16 /text_model/layers.5/self_attn/q_proj/MatMul_myl5_93
|
| 136 |
+
100 kgen BFloat16 __myl_TranMulMean_myl5_94
|
| 137 |
+
101 gemm BFloat16 /text_model/layers.5/self_attn/k_proj/MatMul_myl5_95
|
| 138 |
+
102 kgen BFloat16 __myl_MulMeanAddSqrt_myl5_96
|
| 139 |
+
103 reshape Int64 __mye553436_myl5_97
|
| 140 |
+
104 cast Int64 cast_hvar^553432i64_myl5_98
|
| 141 |
+
105 reshape Int64 __mye553462_myl5_99
|
| 142 |
+
106 cast Int64 cast_hvar^553458i64_myl5_100
|
| 143 |
+
107 kgen BFloat16 __myl_ConcSinCosMoveReshMoveReshAddSqrtDivMulDivMulMulSlicNegSlicConcMulMulAddReshSlicReplReshEtc_myl5_101
|
| 144 |
+
108 gemm BFloat16 /text_model/layers.5/self_attn/v_proj/MatMul_myl5_102
|
| 145 |
+
109 kgen BFloat16 __myl_SlicReplReshMove_myl5_103
|
| 146 |
+
110 kgen FP8 _gemm_mha_v2_myl5_104
|
| 147 |
+
111 kgen FP8 __myl_MoveReshTran_myl5_105
|
| 148 |
+
112 gemm BFloat16 /text_model/layers.5/self_attn/o_proj/MatMul_myl5_106
|
| 149 |
+
113 kgen BFloat16 __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulCastMulCast_myl5_107
|
| 150 |
+
114 gemm BFloat16 /text_model/layers.5/mlp/up_proj/MatMul_myl5_108
|
| 151 |
+
115 gemm BFloat16 /text_model/layers.5/mlp/gate_proj/MatMul_myl5_109
|
| 152 |
+
116 kgen FP8 __myl_MulMulMulAddMulTanhAddMulMulMulCastMulCast_myl5_110
|
| 153 |
+
117 gemm BFloat16 /text_model/layers.5/mlp/down_proj/MatMul_myl5_111
|
| 154 |
+
118 kgen BFloat16 __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulCastMulCast_myl5_112
|
| 155 |
+
119 gemm BFloat16 /text_model/layers.6/self_attn/k_proj/MatMul_myl5_113
|
| 156 |
+
120 kgen BFloat16 __myl_MulMeanAddSqrt_myl5_114
|
| 157 |
+
121 kgen BFloat16 __myl_DivMulMulSlicNegSlicConcMulMulAddReshSlicReplReshTranReshMove_myl5_115
|
| 158 |
+
122 gemm BFloat16 /text_model/layers.6/self_attn/q_proj/MatMul_myl5_116
|
| 159 |
+
123 kgen BFloat16 __myl_TranMulMean_myl5_117
|
| 160 |
+
124 kgen BFloat16 __myl_AddSqrtDivMulMulSlicNegSlicConcMulMulAddReshMove_myl5_118
|
| 161 |
+
125 gemm BFloat16 /text_model/layers.6/self_attn/v_proj/MatMul_myl5_119
|
| 162 |
+
126 kgen BFloat16 __myl_SlicReplReshMove_myl5_120
|
| 163 |
+
127 kgen FP8 _gemm_mha_v2_myl5_121
|
| 164 |
+
128 kgen FP8 __myl_MoveReshTran_myl5_122
|
| 165 |
+
129 gemm BFloat16 /text_model/layers.6/self_attn/o_proj/MatMul_myl5_123
|
| 166 |
+
130 kgen BFloat16 __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulCastMulCast_myl5_124
|
| 167 |
+
131 gemm BFloat16 /text_model/layers.6/mlp/up_proj/MatMul_myl5_125
|
| 168 |
+
132 gemm BFloat16 /text_model/layers.6/mlp/gate_proj/MatMul_myl5_126
|
| 169 |
+
133 kgen FP8 __myl_MulMulMulAddMulTanhAddMulMulMulCastMulCast_myl5_127
|
| 170 |
+
134 gemm BFloat16 /text_model/layers.6/mlp/down_proj/MatMul_myl5_128
|
| 171 |
+
135 kgen BFloat16 __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulCastMulCast_myl5_129
|
| 172 |
+
136 gemm BFloat16 /text_model/layers.7/self_attn/k_proj/MatMul_myl5_130
|
| 173 |
+
137 kgen BFloat16 __myl_MulMeanAddSqrt_myl5_131
|
| 174 |
+
138 kgen BFloat16 __myl_DivMulMulSlicNegSlicConcMulMulAddReshSlicReplReshTranReshMove_myl5_132
|
| 175 |
+
139 gemm BFloat16 /text_model/layers.7/self_attn/q_proj/MatMul_myl5_133
|
| 176 |
+
140 kgen BFloat16 __myl_TranMulMean_myl5_134
|
| 177 |
+
141 kgen BFloat16 __myl_AddSqrtDivMulMulSlicNegSlicConcMulMulAddReshMove_myl5_135
|
| 178 |
+
142 gemm BFloat16 /text_model/layers.7/self_attn/v_proj/MatMul_myl5_136
|
| 179 |
+
143 kgen BFloat16 __myl_SlicReplReshMove_myl5_137
|
| 180 |
+
144 kgen FP8 _gemm_mha_v2_myl5_138
|
| 181 |
+
145 kgen FP8 __myl_MoveReshTran_myl5_139
|
| 182 |
+
146 gemm BFloat16 /text_model/layers.7/self_attn/o_proj/MatMul_myl5_140
|
| 183 |
+
147 kgen BFloat16 __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulCastMulCast_myl5_141
|
| 184 |
+
148 gemm BFloat16 /text_model/layers.7/mlp/up_proj/MatMul_myl5_142
|
| 185 |
+
149 gemm BFloat16 /text_model/layers.7/mlp/gate_proj/MatMul_myl5_143
|
| 186 |
+
150 kgen FP8 __myl_MulMulMulAddMulTanhAddMulMulMulCastMulCast_myl5_144
|
| 187 |
+
151 gemm BFloat16 /text_model/layers.7/mlp/down_proj/MatMul_myl5_145
|
| 188 |
+
152 kgen BFloat16 __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulCastMulCast_myl5_146
|
| 189 |
+
153 gemm BFloat16 /text_model/layers.8/self_attn/k_proj/MatMul_myl5_147
|
| 190 |
+
154 kgen BFloat16 __myl_MulMeanAddSqrt_myl5_148
|
| 191 |
+
155 kgen BFloat16 __myl_DivMulMulSlicNegSlicConcMulMulAddReshSlicReplReshTranReshMove_myl5_149
|
| 192 |
+
156 gemm BFloat16 /text_model/layers.8/self_attn/q_proj/MatMul_myl5_150
|
| 193 |
+
157 kgen BFloat16 __myl_TranMulMean_myl5_151
|
| 194 |
+
158 kgen BFloat16 __myl_AddSqrtDivMulMulSlicNegSlicConcMulMulAddReshMove_myl5_152
|
| 195 |
+
159 gemm BFloat16 /text_model/layers.8/self_attn/v_proj/MatMul_myl5_153
|
| 196 |
+
160 kgen BFloat16 __myl_SlicReplReshMove_myl5_154
|
| 197 |
+
161 kgen FP8 _gemm_mha_v2_myl5_155
|
| 198 |
+
162 kgen FP8 __myl_MoveReshTran_myl5_156
|
| 199 |
+
163 gemm BFloat16 /text_model/layers.8/self_attn/o_proj/MatMul_myl5_157
|
| 200 |
+
164 kgen BFloat16 __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulCastMulCast_myl5_158
|
| 201 |
+
165 gemm BFloat16 /text_model/layers.8/mlp/up_proj/MatMul_myl5_159
|
| 202 |
+
166 gemm BFloat16 /text_model/layers.8/mlp/gate_proj/MatMul_myl5_160
|
| 203 |
+
167 kgen FP8 __myl_MulMulMulAddMulTanhAddMulMulMulCastMulCast_myl5_161
|
| 204 |
+
168 gemm BFloat16 /text_model/layers.8/mlp/down_proj/MatMul_myl5_162
|
| 205 |
+
169 kgen BFloat16 __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulCastMulCast_myl5_163
|
| 206 |
+
170 gemm BFloat16 /text_model/layers.9/self_attn/k_proj/MatMul_myl5_164
|
| 207 |
+
171 kgen BFloat16 __myl_MulMeanAddSqrt_myl5_165
|
| 208 |
+
172 kgen BFloat16 __myl_DivMulMulSlicNegSlicConcMulMulAddReshSlicReplReshTranReshMove_myl5_166
|
| 209 |
+
173 gemm BFloat16 /text_model/layers.9/self_attn/q_proj/MatMul_myl5_167
|
| 210 |
+
174 kgen BFloat16 __myl_TranMulMean_myl5_168
|
| 211 |
+
175 kgen BFloat16 __myl_AddSqrtDivMulMulSlicNegSlicConcMulMulAddReshMove_myl5_169
|
| 212 |
+
176 gemm BFloat16 /text_model/layers.9/self_attn/v_proj/MatMul_myl5_170
|
| 213 |
+
177 kgen BFloat16 __myl_SlicReplReshMove_myl5_171
|
| 214 |
+
178 kgen FP8 _gemm_mha_v2_myl5_172
|
| 215 |
+
179 kgen FP8 __myl_MoveReshTran_myl5_173
|
| 216 |
+
180 gemm BFloat16 /text_model/layers.9/self_attn/o_proj/MatMul_myl5_174
|
| 217 |
+
181 kgen BFloat16 __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulCastMulCast_myl5_175
|
| 218 |
+
182 gemm BFloat16 /text_model/layers.9/mlp/up_proj/MatMul_myl5_176
|
| 219 |
+
183 gemm BFloat16 /text_model/layers.9/mlp/gate_proj/MatMul_myl5_177
|
| 220 |
+
184 kgen FP8 __myl_MulMulMulAddMulTanhAddMulMulMulCastMulCast_myl5_178
|
| 221 |
+
185 gemm BFloat16 /text_model/layers.9/mlp/down_proj/MatMul_myl5_179
|
| 222 |
+
186 kgen BFloat16 __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulCastMulCast_myl5_180
|
| 223 |
+
187 gemm BFloat16 /text_model/layers.10/self_attn/k_proj/MatMul_myl5_181
|
| 224 |
+
188 kgen BFloat16 __myl_MulMeanAddSqrt_myl5_182
|
| 225 |
+
189 kgen BFloat16 __myl_DivMulMulSlicNegSlicConcMulMulAddReshSlicReplReshTranReshMove_myl5_183
|
| 226 |
+
190 gemm BFloat16 /text_model/layers.10/self_attn/q_proj/MatMul_myl5_184
|
| 227 |
+
191 kgen BFloat16 __myl_TranMulMean_myl5_185
|
| 228 |
+
192 kgen BFloat16 __myl_AddSqrtDivMulMulSlicNegSlicConcMulMulAddReshMove_myl5_186
|
| 229 |
+
193 gemm BFloat16 /text_model/layers.10/self_attn/v_proj/MatMul_myl5_187
|
| 230 |
+
194 kgen BFloat16 __myl_SlicReplReshMove_myl5_188
|
| 231 |
+
195 kgen FP8 _gemm_mha_v2_myl5_189
|
| 232 |
+
196 kgen FP8 __myl_MoveReshTran_myl5_190
|
| 233 |
+
197 gemm BFloat16 /text_model/layers.10/self_attn/o_proj/MatMul_myl5_191
|
| 234 |
+
198 kgen BFloat16 __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulCastMulCast_myl5_192
|
| 235 |
+
199 gemm BFloat16 /text_model/layers.10/mlp/up_proj/MatMul_myl5_193
|
| 236 |
+
200 gemm BFloat16 /text_model/layers.10/mlp/gate_proj/MatMul_myl5_194
|
| 237 |
+
201 kgen FP8 __myl_MulMulMulAddMulTanhAddMulMulMulCastMulCast_myl5_195
|
| 238 |
+
202 gemm BFloat16 /text_model/layers.10/mlp/down_proj/MatMul_myl5_196
|
| 239 |
+
203 kgen BFloat16 __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulCastMulCast_myl5_197
|
| 240 |
+
204 gemm BFloat16 /text_model/layers.11/self_attn/k_proj/MatMul_myl5_198
|
| 241 |
+
205 kgen BFloat16 __myl_MulMeanAddSqrt_myl5_199
|
| 242 |
+
206 kgen BFloat16 __myl_DivMulMulSlicNegSlicConcMulMulAddReshSlicReplReshTranReshMove_myl5_200
|
| 243 |
+
207 gemm BFloat16 /text_model/layers.11/self_attn/q_proj/MatMul_myl5_201
|
| 244 |
+
208 kgen BFloat16 __myl_TranMulMean_myl5_202
|
| 245 |
+
209 kgen BFloat16 __myl_AddSqrtDivMulMulSlicNegSlicConcMulMulAddReshMove_myl5_203
|
| 246 |
+
210 gemm BFloat16 /text_model/layers.11/self_attn/v_proj/MatMul_myl5_204
|
| 247 |
+
211 kgen BFloat16 __myl_SlicReplReshMove_myl5_205
|
| 248 |
+
212 kgen FP8 _gemm_mha_v2_myl5_206
|
| 249 |
+
213 kgen FP8 __myl_MoveReshTran_myl5_207
|
| 250 |
+
214 gemm BFloat16 /text_model/layers.11/self_attn/o_proj/MatMul_myl5_208
|
| 251 |
+
215 kgen BFloat16 __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulCastMulCast_myl5_209
|
| 252 |
+
216 gemm BFloat16 /text_model/layers.11/mlp/up_proj/MatMul_myl5_210
|
| 253 |
+
217 gemm BFloat16 /text_model/layers.11/mlp/gate_proj/MatMul_myl5_211
|
| 254 |
+
218 kgen FP8 __myl_MulMulMulAddMulTanhAddMulMulMulCastMulCast_myl5_212
|
| 255 |
+
219 gemm BFloat16 /text_model/layers.11/mlp/down_proj/MatMul_myl5_213
|
| 256 |
+
220 kgen BFloat16 __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulCastMulCast_myl5_214
|
| 257 |
+
221 gemm BFloat16 /text_model/layers.12/self_attn/k_proj/MatMul_myl5_215
|
| 258 |
+
222 kgen BFloat16 __myl_MulMeanAddSqrt_myl5_216
|
| 259 |
+
223 kgen BFloat16 __myl_DivMulMulSlicNegSlicConcMulMulAddReshSlicReplReshTranReshMove_myl5_217
|
| 260 |
+
224 gemm BFloat16 /text_model/layers.12/self_attn/q_proj/MatMul_myl5_218
|
| 261 |
+
225 kgen BFloat16 __myl_TranMulMean_myl5_219
|
| 262 |
+
226 kgen BFloat16 __myl_AddSqrtDivMulMulSlicNegSlicConcMulMulAddReshMove_myl5_220
|
| 263 |
+
227 gemm BFloat16 /text_model/layers.12/self_attn/v_proj/MatMul_myl5_221
|
| 264 |
+
228 kgen BFloat16 __myl_SlicReplReshMove_myl5_222
|
| 265 |
+
229 kgen FP8 _gemm_mha_v2_myl5_223
|
| 266 |
+
230 kgen FP8 __myl_MoveReshTran_myl5_224
|
| 267 |
+
231 gemm BFloat16 /text_model/layers.12/self_attn/o_proj/MatMul_myl5_225
|
| 268 |
+
232 kgen BFloat16 __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulCastMulCast_myl5_226
|
| 269 |
+
233 gemm BFloat16 /text_model/layers.12/mlp/up_proj/MatMul_myl5_227
|
| 270 |
+
234 gemm BFloat16 /text_model/layers.12/mlp/gate_proj/MatMul_myl5_228
|
| 271 |
+
235 kgen FP8 __myl_MulMulMulAddMulTanhAddMulMulMulCastMulCast_myl5_229
|
| 272 |
+
236 gemm BFloat16 /text_model/layers.12/mlp/down_proj/MatMul_myl5_230
|
| 273 |
+
237 kgen BFloat16 __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulCastMulCast_myl5_231
|
| 274 |
+
238 gemm BFloat16 /text_model/layers.13/self_attn/k_proj/MatMul_myl5_232
|
| 275 |
+
239 kgen BFloat16 __myl_MulMeanAddSqrt_myl5_233
|
| 276 |
+
240 kgen BFloat16 __myl_DivMulMulSlicNegSlicConcMulMulAddReshSlicReplReshTranReshMove_myl5_234
|
| 277 |
+
241 gemm BFloat16 /text_model/layers.13/self_attn/q_proj/MatMul_myl5_235
|
| 278 |
+
242 kgen BFloat16 __myl_TranMulMean_myl5_236
|
| 279 |
+
243 kgen BFloat16 __myl_AddSqrtDivMulMulSlicNegSlicConcMulMulAddReshMove_myl5_237
|
| 280 |
+
244 gemm BFloat16 /text_model/layers.13/self_attn/v_proj/MatMul_myl5_238
|
| 281 |
+
245 kgen BFloat16 __myl_SlicReplReshMove_myl5_239
|
| 282 |
+
246 kgen FP8 _gemm_mha_v2_myl5_240
|
| 283 |
+
247 kgen FP8 __myl_MoveReshTran_myl5_241
|
| 284 |
+
248 gemm BFloat16 /text_model/layers.13/self_attn/o_proj/MatMul_myl5_242
|
| 285 |
+
249 kgen BFloat16 __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulCastMulCast_myl5_243
|
| 286 |
+
250 gemm BFloat16 /text_model/layers.13/mlp/up_proj/MatMul_myl5_244
|
| 287 |
+
251 gemm BFloat16 /text_model/layers.13/mlp/gate_proj/MatMul_myl5_245
|
| 288 |
+
252 kgen FP8 __myl_MulMulMulAddMulTanhAddMulMulMulCastMulCast_myl5_246
|
| 289 |
+
253 gemm BFloat16 /text_model/layers.13/mlp/down_proj/MatMul_myl5_247
|
| 290 |
+
254 kgen BFloat16 __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulCastMulCast_myl5_248
|
| 291 |
+
255 gemm BFloat16 /text_model/layers.14/self_attn/k_proj/MatMul_myl5_249
|
| 292 |
+
256 kgen BFloat16 __myl_MulMeanAddSqrt_myl5_250
|
| 293 |
+
257 kgen BFloat16 __myl_DivMulMulSlicNegSlicConcMulMulAddReshSlicReplReshTranReshMove_myl5_251
|
| 294 |
+
258 gemm BFloat16 /text_model/layers.14/self_attn/q_proj/MatMul_myl5_252
|
| 295 |
+
259 kgen BFloat16 __myl_TranMulMean_myl5_253
|
| 296 |
+
260 kgen BFloat16 __myl_AddSqrtDivMulMulSlicNegSlicConcMulMulAddReshMove_myl5_254
|
| 297 |
+
261 gemm BFloat16 /text_model/layers.14/self_attn/v_proj/MatMul_myl5_255
|
| 298 |
+
262 kgen BFloat16 __myl_SlicReplReshMove_myl5_256
|
| 299 |
+
263 kgen FP8 _gemm_mha_v2_myl5_257
|
| 300 |
+
264 kgen FP8 __myl_MoveReshTran_myl5_258
|
| 301 |
+
265 gemm BFloat16 /text_model/layers.14/self_attn/o_proj/MatMul_myl5_259
|
| 302 |
+
266 kgen BFloat16 __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulCastMulCast_myl5_260
|
| 303 |
+
267 gemm BFloat16 /text_model/layers.14/mlp/up_proj/MatMul_myl5_261
|
| 304 |
+
268 gemm BFloat16 /text_model/layers.14/mlp/gate_proj/MatMul_myl5_262
|
| 305 |
+
269 kgen FP8 __myl_MulMulMulAddMulTanhAddMulMulMulCastMulCast_myl5_263
|
| 306 |
+
270 gemm BFloat16 /text_model/layers.14/mlp/down_proj/MatMul_myl5_264
|
| 307 |
+
271 kgen BFloat16 __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulCastMulCast_myl5_265
|
| 308 |
+
272 gemm BFloat16 /text_model/layers.15/self_attn/k_proj/MatMul_myl5_266
|
| 309 |
+
273 kgen BFloat16 __myl_MulMeanAddSqrt_myl5_267
|
| 310 |
+
274 kgen BFloat16 __myl_DivMulMulSlicNegSlicConcMulMulAddReshSlicReplReshTranReshMove_myl5_268
|
| 311 |
+
275 gemm BFloat16 /text_model/layers.15/self_attn/q_proj/MatMul_myl5_269
|
| 312 |
+
276 kgen BFloat16 __myl_TranMulMean_myl5_270
|
| 313 |
+
277 kgen BFloat16 __myl_AddSqrtDivMulMulSlicNegSlicConcMulMulAddReshMove_myl5_271
|
| 314 |
+
278 gemm BFloat16 /text_model/layers.15/self_attn/v_proj/MatMul_myl5_272
|
| 315 |
+
279 kgen BFloat16 __myl_SlicReplReshMove_myl5_273
|
| 316 |
+
280 kgen FP8 _gemm_mha_v2_myl5_274
|
| 317 |
+
281 kgen FP8 __myl_MoveReshTran_myl5_275
|
| 318 |
+
282 gemm BFloat16 /text_model/layers.15/self_attn/o_proj/MatMul_myl5_276
|
| 319 |
+
283 kgen BFloat16 __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulCastMulCast_myl5_277
|
| 320 |
+
284 gemm BFloat16 /text_model/layers.15/mlp/up_proj/MatMul_myl5_278
|
| 321 |
+
285 gemm BFloat16 /text_model/layers.15/mlp/gate_proj/MatMul_myl5_279
|
| 322 |
+
286 kgen FP8 __myl_MulMulMulAddMulTanhAddMulMulMulCastMulCast_myl5_280
|
| 323 |
+
287 gemm BFloat16 /text_model/layers.15/mlp/down_proj/MatMul_myl5_281
|
| 324 |
+
288 kgen BFloat16 __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulCastMulCast_myl5_282
|
| 325 |
+
289 gemm BFloat16 /text_model/layers.16/self_attn/k_proj/MatMul_myl5_283
|
| 326 |
+
290 kgen BFloat16 __myl_MulMeanAddSqrt_myl5_284
|
| 327 |
+
291 kgen BFloat16 __myl_DivMulMulSlicNegSlicConcMulMulAddReshSlicReplReshTranReshMove_myl5_285
|
| 328 |
+
292 gemm BFloat16 /text_model/layers.16/self_attn/q_proj/MatMul_myl5_286
|
| 329 |
+
293 kgen BFloat16 __myl_TranMulMean_myl5_287
|
| 330 |
+
294 kgen BFloat16 __myl_AddSqrtDivMulMulSlicNegSlicConcMulMulAddReshMove_myl5_288
|
| 331 |
+
295 gemm BFloat16 /text_model/layers.16/self_attn/v_proj/MatMul_myl5_289
|
| 332 |
+
296 kgen BFloat16 __myl_SlicReplReshMove_myl5_290
|
| 333 |
+
297 kgen FP8 _gemm_mha_v2_myl5_291
|
| 334 |
+
298 kgen FP8 __myl_MoveReshTran_myl5_292
|
| 335 |
+
299 gemm BFloat16 /text_model/layers.16/self_attn/o_proj/MatMul_myl5_293
|
| 336 |
+
300 kgen BFloat16 __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulCastMulCast_myl5_294
|
| 337 |
+
301 gemm BFloat16 /text_model/layers.16/mlp/up_proj/MatMul_myl5_295
|
| 338 |
+
302 gemm BFloat16 /text_model/layers.16/mlp/gate_proj/MatMul_myl5_296
|
| 339 |
+
303 kgen FP8 __myl_MulMulMulAddMulTanhAddMulMulMulCastMulCast_myl5_297
|
| 340 |
+
304 gemm BFloat16 /text_model/layers.16/mlp/down_proj/MatMul_myl5_298
|
| 341 |
+
305 kgen BFloat16 __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulCastMulCast_myl5_299
|
| 342 |
+
306 gemm BFloat16 /text_model/layers.17/self_attn/k_proj/MatMul_myl5_300
|
| 343 |
+
307 kgen BFloat16 __myl_MulMeanAddSqrt_myl5_301
|
| 344 |
+
308 kgen BFloat16 __myl_DivMulMulSlicNegSlicConcMulMulAddReshSlicReplReshTranReshMove_myl5_302
|
| 345 |
+
309 gemm BFloat16 /text_model/layers.17/self_attn/q_proj/MatMul_myl5_303
|
| 346 |
+
310 kgen BFloat16 __myl_TranMulMean_myl5_304
|
| 347 |
+
311 kgen BFloat16 __myl_AddSqrtDivMulMulSlicNegSlicConcMulMulAddReshMove_myl5_305
|
| 348 |
+
312 gemm BFloat16 /text_model/layers.17/self_attn/v_proj/MatMul_myl5_306
|
| 349 |
+
313 kgen BFloat16 __myl_SlicReplReshMove_myl5_307
|
| 350 |
+
314 kgen FP8 _gemm_mha_v2_myl5_308
|
| 351 |
+
315 kgen FP8 __myl_MoveReshTran_myl5_309
|
| 352 |
+
316 gemm BFloat16 /text_model/layers.17/self_attn/o_proj/MatMul_myl5_310
|
| 353 |
+
317 kgen BFloat16 __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulCastMulCast_myl5_311
|
| 354 |
+
318 gemm BFloat16 /text_model/layers.17/mlp/up_proj/MatMul_myl5_312
|
| 355 |
+
319 gemm BFloat16 /text_model/layers.17/mlp/gate_proj/MatMul_myl5_313
|
| 356 |
+
320 kgen FP8 __myl_MulMulMulAddMulTanhAddMulMulMulCastMulCast_myl5_314
|
| 357 |
+
321 gemm BFloat16 /text_model/layers.17/mlp/down_proj/MatMul_myl5_315
|
| 358 |
+
322 kgen BFloat16 __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulCastMulCast_myl5_316
|
| 359 |
+
323 gemm BFloat16 /text_model/layers.18/self_attn/k_proj/MatMul_myl5_317
|
| 360 |
+
324 kgen BFloat16 __myl_MulMeanAddSqrt_myl5_318
|
| 361 |
+
325 kgen BFloat16 __myl_DivMulMulSlicNegSlicConcMulMulAddReshSlicReplReshTranReshMove_myl5_319
|
| 362 |
+
326 gemm BFloat16 /text_model/layers.18/self_attn/q_proj/MatMul_myl5_320
|
| 363 |
+
327 kgen BFloat16 __myl_TranMulMean_myl5_321
|
| 364 |
+
328 kgen BFloat16 __myl_AddSqrtDivMulMulSlicNegSlicConcMulMulAddReshMove_myl5_322
|
| 365 |
+
329 gemm BFloat16 /text_model/layers.18/self_attn/v_proj/MatMul_myl5_323
|
| 366 |
+
330 kgen BFloat16 __myl_SlicReplReshMove_myl5_324
|
| 367 |
+
331 kgen FP8 _gemm_mha_v2_myl5_325
|
| 368 |
+
332 kgen FP8 __myl_MoveReshTran_myl5_326
|
| 369 |
+
333 gemm BFloat16 /text_model/layers.18/self_attn/o_proj/MatMul_myl5_327
|
| 370 |
+
334 kgen BFloat16 __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulCastMulCast_myl5_328
|
| 371 |
+
335 gemm BFloat16 /text_model/layers.18/mlp/up_proj/MatMul_myl5_329
|
| 372 |
+
336 gemm BFloat16 /text_model/layers.18/mlp/gate_proj/MatMul_myl5_330
|
| 373 |
+
337 kgen FP8 __myl_MulMulMulAddMulTanhAddMulMulMulCastMulCast_myl5_331
|
| 374 |
+
338 gemm BFloat16 /text_model/layers.18/mlp/down_proj/MatMul_myl5_332
|
| 375 |
+
339 kgen BFloat16 __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulCastMulCast_myl5_333
|
| 376 |
+
340 gemm BFloat16 /text_model/layers.19/self_attn/k_proj/MatMul_myl5_334
|
| 377 |
+
341 kgen BFloat16 __myl_MulMeanAddSqrt_myl5_335
|
| 378 |
+
342 kgen BFloat16 __myl_DivMulMulSlicNegSlicConcMulMulAddReshSlicReplReshTranReshMove_myl5_336
|
| 379 |
+
343 gemm BFloat16 /text_model/layers.19/self_attn/q_proj/MatMul_myl5_337
|
| 380 |
+
344 kgen BFloat16 __myl_TranMulMean_myl5_338
|
| 381 |
+
345 kgen BFloat16 __myl_AddSqrtDivMulMulSlicNegSlicConcMulMulAddReshMove_myl5_339
|
| 382 |
+
346 gemm BFloat16 /text_model/layers.19/self_attn/v_proj/MatMul_myl5_340
|
| 383 |
+
347 kgen BFloat16 __myl_SlicReplReshMove_myl5_341
|
| 384 |
+
348 kgen FP8 _gemm_mha_v2_myl5_342
|
| 385 |
+
349 kgen FP8 __myl_MoveReshTran_myl5_343
|
| 386 |
+
350 gemm BFloat16 /text_model/layers.19/self_attn/o_proj/MatMul_myl5_344
|
| 387 |
+
351 kgen BFloat16 __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulCastMulCast_myl5_345
|
| 388 |
+
352 gemm BFloat16 /text_model/layers.19/mlp/up_proj/MatMul_myl5_346
|
| 389 |
+
353 gemm BFloat16 /text_model/layers.19/mlp/gate_proj/MatMul_myl5_347
|
| 390 |
+
354 kgen FP8 __myl_MulMulMulAddMulTanhAddMulMulMulCastMulCast_myl5_348
|
| 391 |
+
355 gemm BFloat16 /text_model/layers.19/mlp/down_proj/MatMul_myl5_349
|
| 392 |
+
356 kgen BFloat16 __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulCastMulCast_myl5_350
|
| 393 |
+
357 gemm BFloat16 /text_model/layers.20/self_attn/k_proj/MatMul_myl5_351
|
| 394 |
+
358 kgen BFloat16 __myl_MulMeanAddSqrt_myl5_352
|
| 395 |
+
359 kgen BFloat16 __myl_DivMulMulSlicNegSlicConcMulMulAddReshSlicReplReshTranReshMove_myl5_353
|
| 396 |
+
360 gemm BFloat16 /text_model/layers.20/self_attn/q_proj/MatMul_myl5_354
|
| 397 |
+
361 kgen BFloat16 __myl_TranMulMean_myl5_355
|
| 398 |
+
362 kgen BFloat16 __myl_AddSqrtDivMulMulSlicNegSlicConcMulMulAddReshMove_myl5_356
|
| 399 |
+
363 gemm BFloat16 /text_model/layers.20/self_attn/v_proj/MatMul_myl5_357
|
| 400 |
+
364 kgen BFloat16 __myl_SlicReplReshMove_myl5_358
|
| 401 |
+
365 kgen FP8 _gemm_mha_v2_myl5_359
|
| 402 |
+
366 kgen FP8 __myl_MoveReshTran_myl5_360
|
| 403 |
+
367 gemm BFloat16 /text_model/layers.20/self_attn/o_proj/MatMul_myl5_361
|
| 404 |
+
368 kgen BFloat16 __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulCastMulCast_myl5_362
|
| 405 |
+
369 gemm BFloat16 /text_model/layers.20/mlp/up_proj/MatMul_myl5_363
|
| 406 |
+
370 gemm BFloat16 /text_model/layers.20/mlp/gate_proj/MatMul_myl5_364
|
| 407 |
+
371 kgen FP8 __myl_MulMulMulAddMulTanhAddMulMulMulCastMulCast_myl5_365
|
| 408 |
+
372 gemm BFloat16 /text_model/layers.20/mlp/down_proj/MatMul_myl5_366
|
| 409 |
+
373 kgen BFloat16 __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulCastMulCast_myl5_367
|
| 410 |
+
374 gemm BFloat16 /text_model/layers.21/self_attn/k_proj/MatMul_myl5_368
|
| 411 |
+
375 kgen BFloat16 __myl_MulMeanAddSqrt_myl5_369
|
| 412 |
+
376 kgen BFloat16 __myl_DivMulMulSlicNegSlicConcMulMulAddReshSlicReplReshTranReshMove_myl5_370
|
| 413 |
+
377 gemm BFloat16 /text_model/layers.21/self_attn/q_proj/MatMul_myl5_371
|
| 414 |
+
378 kgen BFloat16 __myl_TranMulMean_myl5_372
|
| 415 |
+
379 kgen BFloat16 __myl_AddSqrtDivMulMulSlicNegSlicConcMulMulAddReshMove_myl5_373
|
| 416 |
+
380 gemm BFloat16 /text_model/layers.21/self_attn/v_proj/MatMul_myl5_374
|
| 417 |
+
381 kgen BFloat16 __myl_SlicReplReshMove_myl5_375
|
| 418 |
+
382 kgen FP8 _gemm_mha_v2_myl5_376
|
| 419 |
+
383 kgen FP8 __myl_MoveReshTran_myl5_377
|
| 420 |
+
384 gemm BFloat16 /text_model/layers.21/self_attn/o_proj/MatMul_myl5_378
|
| 421 |
+
385 kgen BFloat16 __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulCastMulCast_myl5_379
|
| 422 |
+
386 gemm BFloat16 /text_model/layers.21/mlp/up_proj/MatMul_myl5_380
|
| 423 |
+
387 gemm BFloat16 /text_model/layers.21/mlp/gate_proj/MatMul_myl5_381
|
| 424 |
+
388 kgen FP8 __myl_MulMulMulAddMulTanhAddMulMulMulCastMulCast_myl5_382
|
| 425 |
+
389 gemm BFloat16 /text_model/layers.21/mlp/down_proj/MatMul_myl5_383
|
| 426 |
+
390 kgen BFloat16 __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulCastMulCast_myl5_384
|
| 427 |
+
391 gemm BFloat16 /text_model/layers.22/self_attn/k_proj/MatMul_myl5_385
|
| 428 |
+
392 kgen BFloat16 __myl_MulMeanAddSqrt_myl5_386
|
| 429 |
+
393 kgen BFloat16 __myl_DivMulMulSlicNegSlicConcMulMulAddReshSlicReplReshTranReshMove_myl5_387
|
| 430 |
+
394 gemm BFloat16 /text_model/layers.22/self_attn/q_proj/MatMul_myl5_388
|
| 431 |
+
395 kgen BFloat16 __myl_TranMulMean_myl5_389
|
| 432 |
+
396 kgen BFloat16 __myl_AddSqrtDivMulMulSlicNegSlicConcMulMulAddReshMove_myl5_390
|
| 433 |
+
397 gemm BFloat16 /text_model/layers.22/self_attn/v_proj/MatMul_myl5_391
|
| 434 |
+
398 kgen BFloat16 __myl_SlicReplReshMove_myl5_392
|
| 435 |
+
399 kgen FP8 _gemm_mha_v2_myl5_393
|
| 436 |
+
400 kgen FP8 __myl_MoveReshTran_myl5_394
|
| 437 |
+
401 gemm BFloat16 /text_model/layers.22/self_attn/o_proj/MatMul_myl5_395
|
| 438 |
+
402 kgen BFloat16 __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulCastMulCast_myl5_396
|
| 439 |
+
403 gemm BFloat16 /text_model/layers.22/mlp/up_proj/MatMul_myl5_397
|
| 440 |
+
404 gemm BFloat16 /text_model/layers.22/mlp/gate_proj/MatMul_myl5_398
|
| 441 |
+
405 kgen FP8 __myl_MulMulMulAddMulTanhAddMulMulMulCastMulCast_myl5_399
|
| 442 |
+
406 gemm BFloat16 /text_model/layers.22/mlp/down_proj/MatMul_myl5_400
|
| 443 |
+
407 kgen BFloat16 __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulCastMulCast_myl5_401
|
| 444 |
+
408 gemm BFloat16 /text_model/layers.23/self_attn/k_proj/MatMul_myl5_402
|
| 445 |
+
409 kgen BFloat16 __myl_MulMeanAddSqrt_myl5_403
|
| 446 |
+
410 kgen BFloat16 __myl_DivMulMulSlicNegSlicConcMulMulAddReshSlicReplReshTranReshMove_myl5_404
|
| 447 |
+
411 gemm BFloat16 /text_model/layers.23/self_attn/q_proj/MatMul_myl5_405
|
| 448 |
+
412 kgen BFloat16 __myl_TranMulMean_myl5_406
|
| 449 |
+
413 kgen BFloat16 __myl_AddSqrtDivMulMulSlicNegSlicConcMulMulAddReshMove_myl5_407
|
| 450 |
+
414 gemm BFloat16 /text_model/layers.23/self_attn/v_proj/MatMul_myl5_408
|
| 451 |
+
415 kgen BFloat16 __myl_SlicReplReshMove_myl5_409
|
| 452 |
+
416 kgen FP8 _gemm_mha_v2_myl5_410
|
| 453 |
+
417 kgen FP8 __myl_MoveReshTran_myl5_411
|
| 454 |
+
418 gemm BFloat16 /text_model/layers.23/self_attn/o_proj/MatMul_myl5_412
|
| 455 |
+
419 kgen BFloat16 __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulCastMulCast_myl5_413
|
| 456 |
+
420 gemm BFloat16 /text_model/layers.23/mlp/up_proj/MatMul_myl5_414
|
| 457 |
+
421 gemm BFloat16 /text_model/layers.23/mlp/gate_proj/MatMul_myl5_415
|
| 458 |
+
422 kgen FP8 __myl_MulMulMulAddMulTanhAddMulMulMulCastMulCast_myl5_416
|
| 459 |
+
423 gemm BFloat16 /text_model/layers.23/mlp/down_proj/MatMul_myl5_417
|
| 460 |
+
424 kgen BFloat16 __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulCastMulCast_myl5_418
|
| 461 |
+
425 gemm BFloat16 /text_model/layers.24/self_attn/k_proj/MatMul_myl5_419
|
| 462 |
+
426 kgen BFloat16 __myl_MulMeanAddSqrt_myl5_420
|
| 463 |
+
427 kgen BFloat16 __myl_DivMulMulSlicNegSlicConcMulMulAddReshSlicReplReshTranReshMove_myl5_421
|
| 464 |
+
428 gemm BFloat16 /text_model/layers.24/self_attn/q_proj/MatMul_myl5_422
|
| 465 |
+
429 kgen BFloat16 __myl_TranMulMean_myl5_423
|
| 466 |
+
430 kgen BFloat16 __myl_AddSqrtDivMulMulSlicNegSlicConcMulMulAddReshMove_myl5_424
|
| 467 |
+
431 gemm BFloat16 /text_model/layers.24/self_attn/v_proj/MatMul_myl5_425
|
| 468 |
+
432 kgen BFloat16 __myl_SlicReplReshMove_myl5_426
|
| 469 |
+
433 kgen FP8 _gemm_mha_v2_myl5_427
|
| 470 |
+
434 kgen FP8 __myl_MoveReshTran_myl5_428
|
| 471 |
+
435 gemm BFloat16 /text_model/layers.24/self_attn/o_proj/MatMul_myl5_429
|
| 472 |
+
436 kgen BFloat16 __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulCastMulCast_myl5_430
|
| 473 |
+
437 gemm BFloat16 /text_model/layers.24/mlp/up_proj/MatMul_myl5_431
|
| 474 |
+
438 gemm BFloat16 /text_model/layers.24/mlp/gate_proj/MatMul_myl5_432
|
| 475 |
+
439 kgen FP8 __myl_MulMulMulAddMulTanhAddMulMulMulCastMulCast_myl5_433
|
| 476 |
+
440 gemm BFloat16 /text_model/layers.24/mlp/down_proj/MatMul_myl5_434
|
| 477 |
+
441 kgen BFloat16 __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulCastMulCast_myl5_435
|
| 478 |
+
442 gemm BFloat16 /text_model/layers.25/self_attn/k_proj/MatMul_myl5_436
|
| 479 |
+
443 kgen BFloat16 __myl_MulMeanAddSqrt_myl5_437
|
| 480 |
+
444 kgen BFloat16 __myl_DivMulMulSlicNegSlicConcMulMulAddReshSlicReplReshTranReshMove_myl5_438
|
| 481 |
+
445 gemm BFloat16 /text_model/layers.25/self_attn/q_proj/MatMul_myl5_439
|
| 482 |
+
446 kgen BFloat16 __myl_TranMulMean_myl5_440
|
| 483 |
+
447 kgen BFloat16 __myl_AddSqrtDivMulMulSlicNegSlicConcMulMulAddReshMove_myl5_441
|
| 484 |
+
448 gemm BFloat16 /text_model/layers.25/self_attn/v_proj/MatMul_myl5_442
|
| 485 |
+
449 kgen BFloat16 __myl_SlicReplReshMove_myl5_443
|
| 486 |
+
450 kgen FP8 _gemm_mha_v2_myl5_444
|
| 487 |
+
451 kgen FP8 __myl_MoveReshTran_myl5_445
|
| 488 |
+
452 gemm BFloat16 /text_model/layers.25/self_attn/o_proj/MatMul_myl5_446
|
| 489 |
+
453 kgen BFloat16 __myl_MulMeanAddSqrtDivMulMulAddMulMeanAddSqrtDivMulCastMulCast_myl5_447
|
| 490 |
+
454 gemm BFloat16 /text_model/layers.25/mlp/up_proj/MatMul_myl5_448
|
| 491 |
+
455 gemm BFloat16 /text_model/layers.25/mlp/gate_proj/MatMul_myl5_449
|
| 492 |
+
456 kgen FP8 __myl_MulMulMulAddMulTanhAddMulMulMulCastMulCast_myl5_450
|
| 493 |
+
457 gemm BFloat16 /text_model/layers.25/mlp/down_proj/MatMul_myl5_451
|
| 494 |
+
458 kgen BFloat16 __myl_MulMeanAddSqrtDivMulMulAddMulMean_myl5_452
|
| 495 |
+
459 kgen BFloat16 __myl_ReshCastSum_myl5_453
|
| 496 |
+
460 kgen FP8 __myl_ReshMaxMinAddSqrtDivMulMulReshMulSumReshDivMulCastMulCast_myl5_454
|
| 497 |
+
461 gemm BFloat16 /classifier/Gemm_myl5_455
|
| 498 |
+
462 kgen BFloat16 __myl_Slic_myl5_456
|
| 499 |
+
463 shape_call - __mye26_0_myl6_0
|
| 500 |
+
464 kgen Float __myl_Cast_myl6_1
|
| 501 |
+
465 TrainStation - [trainStation3]
|