Skip to content

Commit ad12b79

Browse files
rosenrodtrosenrodt
authored
[https://nvbugs/5661741][fix] Fix accuracy issue in TRTLLM MoE introduced in #9377 (#9999)
Signed-off-by: Anthony Chang <[email protected]>
1 parent 9eb5a22 commit ad12b79

File tree

113 files changed

+293
-293
lines changed

Some content is hidden

Large Commits have some content hidden by default. Use the searchbox below for content that may be hidden.

113 files changed

+293
-293
lines changed

‎cpp/tensorrt_llm/kernels/trtllmGenKernels/batchedGemm/trtllmGen_bmm_export/KernelMetaInfo.h‎

Lines changed: 113 additions & 113 deletions
Large diffs are not rendered by default.

‎cpp/tensorrt_llm/kernels/trtllmGenKernels/batchedGemm/trtllmGen_bmm_export/cubins/Bmm_E2m1_E2m1E2m1_Fp32_t128x128x256_s6_et128x32_m256x128x64_cga2x1x1_16dp256b_rM_TN_transOut_schedP2x1x2x3_biasM_bN_tma_tmaOpt_clmp_geGlu_dynBatch_sm100f_cubin.cpp‎

Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -1,3 +1,3 @@
11
version https://git-lfs.github.com/spec/v1
2-
oid sha256:aeb5cd123bafc5a63efa28435422b132354c7ff3d3010fbbcada752a54b7eb00
3-
size 1157648
2+
oid sha256:25bd7f21415ede7acd0d96423d6ffaf55945ee0d8231996f82d821970cd49128
3+
size 1160804

‎cpp/tensorrt_llm/kernels/trtllmGenKernels/batchedGemm/trtllmGen_bmm_export/cubins/Bmm_E2m1_E2m1E2m1_Fp32_t128x128x256_s6_et128x32_m256x128x64_cga2x1x1_16dp256b_rM_TN_transOut_schedP2x1x2x3_biasM_bN_tma_tmaOpt_clmp_swiGlu_dynBatch_sm100f_cubin.cpp‎

Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -1,3 +1,3 @@
11
version https://git-lfs.github.com/spec/v1
2-
oid sha256:057f760b9976712414d4f2d48114d2855bbc114f9e275c9ce02e4e79344bcb6a
3-
size 1134858
2+
oid sha256:6544003fbf6ebd0322ef6fff4b888607a8b4af20d5d79e2abc29273c271f3faa
3+
size 1138064

‎cpp/tensorrt_llm/kernels/trtllmGenKernels/batchedGemm/trtllmGen_bmm_export/cubins/Bmm_E2m1_E2m1E2m1_Fp32_t128x128x256u2_s6_et128x32_m256x128x64_cga2x1x1_16dp256b_rM_TN_transOut_schedP2x1x2x3_biasM_bN_tma_tmaOpt_clmp_geGlu_dynBatch_sm100f_cubin.cpp‎

Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -1,3 +1,3 @@
11
version https://git-lfs.github.com/spec/v1
2-
oid sha256:af6feb5e3e7dcdbf84b175e3f95e3c71679ec81fbee1c08add0437282a7ba567
3-
size 1172156
2+
oid sha256:27ab3fd0dde9b0b693d9d77603d1b5639651fefa71f722f94dd96faf4efb93be
3+
size 1175362

‎cpp/tensorrt_llm/kernels/trtllmGenKernels/batchedGemm/trtllmGen_bmm_export/cubins/Bmm_E2m1_E2m1E2m1_Fp32_t128x128x256u2_s6_et128x32_m256x128x64_cga2x1x1_16dp256b_rM_TN_transOut_schedP2x1x2x3_biasM_bN_tma_tmaOpt_clmp_swiGlu_dynBatch_sm100f_cubin.cpp‎

Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -1,3 +1,3 @@
11
version https://git-lfs.github.com/spec/v1
2-
oid sha256:a01cd18f1c60dd8cbdcb76351f3d70a6bbde4596d1740df688664503aa963a2c
3-
size 1150204
2+
oid sha256:603bb29603b9870df3f9e738271289e90691b585d09969e28c9fac8733cee25a
3+
size 1153362

‎cpp/tensorrt_llm/kernels/trtllmGenKernels/batchedGemm/trtllmGen_bmm_export/cubins/Bmm_E2m1_E2m1E2m1_Fp32_t128x16x256_s9_et128x16_m128x16x64_cga1x1x1_16dp256b_rM_TN_transOut_schedP2x1x2x3_biasM_bN_tma_tmaOpt_clmp_geGlu_dynBatch_sm100f_cubin.cpp‎

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -1,3 +1,3 @@
11
version https://git-lfs.github.com/spec/v1
2-
oid sha256:48ece823140284fa80a98866bc9972a75a192c0a5e18dd49a408391348f220b8
2+
oid sha256:a0cc765c518f5a3ebc0ca533d702a6637d2f307ee8d805b73130fe6f5596d07f
33
size 632096

‎cpp/tensorrt_llm/kernels/trtllmGenKernels/batchedGemm/trtllmGen_bmm_export/cubins/Bmm_E2m1_E2m1E2m1_Fp32_t128x16x256_s9_et128x16_m128x16x64_cga1x1x1_16dp256b_rM_TN_transOut_schedP2x1x2x3_biasM_bN_tma_tmaOpt_clmp_swiGlu_dynBatch_sm100f_cubin.cpp‎

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -1,3 +1,3 @@
11
version https://git-lfs.github.com/spec/v1
2-
oid sha256:19dd3971002732fcfa0dac8f2ee9921feeb7493ec7f3612c66f6e2a2b302b0ed
2+
oid sha256:b8af522d4aa4ce9353782d9f0ae15f4bb0751a03de17138cf284a95acd2b61ee
33
size 629878

‎cpp/tensorrt_llm/kernels/trtllmGenKernels/batchedGemm/trtllmGen_bmm_export/cubins/Bmm_E2m1_E2m1E2m1_Fp32_t128x16x256_s9_et128x16_m128x16x64_cga1x1x1_16dp256b_rM_TN_transOut_schedS_biasM_bN_tma_tmaOpt_clmp_geGlu_dynBatch_sm100f_cubin.cpp‎

Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -1,3 +1,3 @@
11
version https://git-lfs.github.com/spec/v1
2-
oid sha256:70c41810e1221972acdc6cacc8b58fc8a82cb50b3f004fee7dc36879b303ba6a
3-
size 513977
2+
oid sha256:c8b755c5f0145d93a223887fcd961eb2d3ec6907c79c6f3ea3568386668f50d8
3+
size 514765

‎cpp/tensorrt_llm/kernels/trtllmGenKernels/batchedGemm/trtllmGen_bmm_export/cubins/Bmm_E2m1_E2m1E2m1_Fp32_t128x16x256_s9_et128x16_m128x16x64_cga1x1x1_16dp256b_rM_TN_transOut_schedS_biasM_bN_tma_tmaOpt_clmp_swiGlu_dynBatch_sm100f_cubin.cpp‎

Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -1,3 +1,3 @@
11
version https://git-lfs.github.com/spec/v1
2-
oid sha256:0df6169b607713ea44bc5b19d7c6c6a2230bd9b99c28dd444c48a8080ada3853
3-
size 511759
2+
oid sha256:dedc17afe8442c3dbf7ea8f883a19d0cdcf85a92161991b53f0f66f63f988814
3+
size 512597

‎cpp/tensorrt_llm/kernels/trtllmGenKernels/batchedGemm/trtllmGen_bmm_export/cubins/Bmm_E2m1_E2m1E2m1_Fp32_t128x16x256u2_s9_et128x16_m128x16x64_cga1x1x1_16dp256b_rM_TN_transOut_schedP2x1x2x3_biasM_bN_tma_tmaOpt_clmp_geGlu_dynBatch_sm100f_cubin.cpp‎

Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -1,3 +1,3 @@
11
version https://git-lfs.github.com/spec/v1
2-
oid sha256:0451e00277a31f78c563b3e4f613c3b0174dc0b1b6657e9a6ab8df2758d33c59
3-
size 652968
2+
oid sha256:f29a3484448fdb88d692b59c3aefb8763c07aaac980e4426f338df973171470f
3+
size 653016

0 commit comments

Comments
 (0)