Skip to content

Actions: NVIDIA/TransformerEngine

Actions

Documentation

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
docs.yml
2,500+ workflow runs
2,500+ workflow runs

Filter by Event

⌥ + click/return to exclude

Filter by Status

⌥ + click/return to exclude

Filter by Branch

⌥ + click/return to exclude

Filter by Actor

⌥ + click/return to exclude
Stabilize FlashAttention CI environments Documentation #19376: Pull request #3341 synchronize by sudhakarsingh27
1m 31s sudhakarsingh27:sudhakars/gate-fa4-sm8x-test-utils
sudhakarsingh27:sudhakars/gate-fa4-sm8x-test-utils
1m 31s
[Pytorch] Enable TE Op to consume extra_outputs from a previously run Op in TE Sequential Documentation #19375: Pull request #3320 synchronize by vthumbe1503
1m 33s vthumbe1503:enable_extra_out_consumption
vthumbe1503:enable_extra_out_consumption
1m 33s
[Pytorch] Optimize CPU overhead for NCCL EP eager mode Documentation #19374: Pull request #3378 opened by YangFei1990
1m 37s YangFei1990:ncclep_cpu_overhead
YangFei1990:ncclep_cpu_overhead
1m 37s
Prefer Flash Attn 2 instead of Fused Attn for THD dropout on Blackwell Documentation #19373: Pull request #3313 synchronize by bzantium
Action required bzantium:log-thd-dropout-composite-engine
bzantium:log-thd-dropout-composite-engine
Action required
Prefer Flash Attn 2 instead of Fused Attn for THD dropout on Blackwell Documentation #19372: Pull request #3313 synchronize by bzantium
Action required bzantium:log-thd-dropout-composite-engine
bzantium:log-thd-dropout-composite-engine
Action required
fix: remove redundant self-assignment out_ = out_ Documentation #19371: Pull request #3367 synchronize by pre-commit-ci Bot
Action required andrewwhitecdw:codequality/run-attention-with-cp-remove-redundant-self-assignment-out-out
andrewwhitecdw:codequality/run-attention-with-cp-remove-redundant-self-assignment-out-out
Action required
fix: add missing NVTE_BHSD to qkv format to_string Documentation #19370: Pull request #3362 synchronize by pre-commit-ci Bot
Action required andrewwhitecdw:bugfix/fused-attn-add-missing-nvte-bhsd-to-qkv-format-to
andrewwhitecdw:bugfix/fused-attn-add-missing-nvte-bhsd-to-qkv-format-to
Action required
fix: remove redundant second export of XLA_FLAGS in L0 Documentation #19369: Pull request #3377 opened by andrewwhitecdw
Action required andrewwhitecdw:codequality/test-remove-redundant-second-export-of-xla
andrewwhitecdw:codequality/test-remove-redundant-second-export-of-xla
Action required
fix: duplicate JUnit XML filename overwrites encoder test results Documentation #19368: Pull request #3376 opened by andrewwhitecdw
Action required andrewwhitecdw:bugfix/test-duplicate-junit-xml-filename
andrewwhitecdw:bugfix/test-duplicate-junit-xml-filename
Action required
fix: L0 placeholder shapes are invalid empty tuple/list Documentation #19367: Pull request #3375 opened by andrewwhitecdw
Action required andrewwhitecdw:bugfix/test-distributed-fused-attn-l0-placeholder-shapes-are-invalid
andrewwhitecdw:bugfix/test-distributed-fused-attn-l0-placeholder-shapes-are-invalid
Action required
fix: allgather striped window_size (5,0) id says (8,0) Documentation #19366: Pull request #3374 opened by andrewwhitecdw
Action required andrewwhitecdw:bugfix/test-distributed-fused-attn-allgather-striped-window-size-5-0-id
andrewwhitecdw:bugfix/test-distributed-fused-attn-allgather-striped-window-size-5-0-id
Action required
fix: SBHD reorder skip uses original shape instead of swapped tensor Documentation #19365: Pull request #3373 opened by andrewwhitecdw
Action required andrewwhitecdw:bugfix/test-distributed-fused-attn-sbhd-reorder-skip-uses-original-shape
andrewwhitecdw:bugfix/test-distributed-fused-attn-sbhd-reorder-skip-uses-original-shape
Action required
fix: comment typo 'mas' -> 'mask' in two TODOs Documentation #19364: Pull request #3372 opened by andrewwhitecdw
Action required andrewwhitecdw:codequality/test-fused-attn-comment-typo-mas-mask-in-two-todos
andrewwhitecdw:codequality/test-fused-attn-comment-typo-mas-mask-in-two-todos
Action required
fix: make_mask uses deprecated jnp.bool instead of jnp.bool_ Documentation #19363: Pull request #3371 opened by andrewwhitecdw
Action required andrewwhitecdw:bugfix/test-fused-attn-make-mask-uses-deprecated-jnp-bool
andrewwhitecdw:bugfix/test-fused-attn-make-mask-uses-deprecated-jnp-bool
Action required
fix: typo and unnecessary f-string in BRCM skip message Documentation #19362: Pull request #3370 opened by andrewwhitecdw
Action required andrewwhitecdw:codequality/test-fused-attn-typo-and-unnecessary-f-string-in-brcm
andrewwhitecdw:codequality/test-fused-attn-typo-and-unnecessary-f-string-in-brcm
Action required
fix: Optional segment position annotations in mask helpers Documentation #19361: Pull request #3369 opened by andrewwhitecdw
Action required andrewwhitecdw:codequality/test-fused-attn-optional-segment-position-annotations
andrewwhitecdw:codequality/test-fused-attn-optional-segment-position-annotations
Action required
fix: Unreachable backend check after earlier backend skip Documentation #19360: Pull request #3368 opened by andrewwhitecdw
Action required andrewwhitecdw:codequality/test-fused-attn-unreachable-backend-check-after
andrewwhitecdw:codequality/test-fused-attn-unreachable-backend-check-after
Action required
fix: remove redundant self-assignment out_ = out_ Documentation #19359: Pull request #3367 opened by andrewwhitecdw
Action required andrewwhitecdw:codequality/run-attention-with-cp-remove-redundant-self-assignment-out-out
andrewwhitecdw:codequality/run-attention-with-cp-remove-redundant-self-assignment-out-out
Action required
fix: fp8 scaling_mode is not validated before use Documentation #19358: Pull request #3366 opened by andrewwhitecdw
Action required andrewwhitecdw:codequality/run-attention-with-cp-fp8-scaling-mode-is-not-validated
andrewwhitecdw:codequality/run-attention-with-cp-fp8-scaling-mode-is-not-validated
Action required
fix: simplify qkv_format checks using a membership tuple Documentation #19357: Pull request #3365 opened by andrewwhitecdw
Action required andrewwhitecdw:codequality/run-attention-with-cp-simplify-qkv-format-checks-using-a
andrewwhitecdw:codequality/run-attention-with-cp-simplify-qkv-format-checks-using-a
Action required
fix: remove unused import time Documentation #19356: Pull request #3364 opened by andrewwhitecdw
Action required andrewwhitecdw:codequality/test-attention-with-cp-remove-unused-import-time
andrewwhitecdw:codequality/test-attention-with-cp-remove-unused-import-time
Action required
fix: simplify redundant softmax version condition Documentation #19355: Pull request #3363 opened by andrewwhitecdw
Action required andrewwhitecdw:codequality/fused-attn-simplify-redundant-softmax-version
andrewwhitecdw:codequality/fused-attn-simplify-redundant-softmax-version
Action required
fix: add missing NVTE_BHSD to qkv format to_string Documentation #19354: Pull request #3362 opened by andrewwhitecdw
Action required andrewwhitecdw:bugfix/fused-attn-add-missing-nvte-bhsd-to-qkv-format-to
andrewwhitecdw:bugfix/fused-attn-add-missing-nvte-bhsd-to-qkv-format-to
Action required
fix: add missing NVTE_BHSD_BHSD_BHSD to layout to_string Documentation #19353: Pull request #3361 opened by andrewwhitecdw
Action required andrewwhitecdw:bugfix/fused-attn-add-missing-nvte-bhsd-bhsd-bhsd-to
andrewwhitecdw:bugfix/fused-attn-add-missing-nvte-bhsd-bhsd-bhsd-to
Action required
Add 2d quant for mxfp8 Documentation #19352: Pull request #2634 synchronize by ptrendx
3m 56s kunlunl:2d_mxfp8
kunlunl:2d_mxfp8
3m 56s