Liger
Jensen-Shannon divergence loss · vocab = 65536 · tokens = 8192 · fp32 · run 01a016ae-6ade…
●passedReported evidence
Not re-observed recently.
Primary measurement
43.5msmedian
Rank 2 in its comparison group · source-native comparison · observed 2024-10-02
ms 20 43.47438049316406 · ms 50 43.49095916748047 · ms 80 43.50754165649414
Identity
implementationLiger
projectLiger-Kernel
revisionunknown
workloadvocab = 65536 · tokens = 8192 · fp32
comparison keysha256:599a1fdb30a2d2ce…
sourceLiger-Kernel benchmarks
external idjsd/forward/liger/nvidia-h100-pcie/tokens8192-vocab65536-fp32/2024-10-02-16-21-37
sha256:65d9c30cd7c689fd16adb0d…
Correctness
Marked passed by the source; the correctness policy was not published.
Workload
vocab65536
tokens8192
student_log_probsfp32 [8192, 65536]
teacher_log_probsfp32 [8192, 65536]
definition comparatornot_asserted
Measurements
latency · median43.5 ms
latency · p2043.5 ms
latency · p8043.5 ms
Protocol
harnessLiger-Kernel benchmark scripts
timercuda_events
primaryStatisticmedian
comparabilityFamilyliger_kernel_bench
Environment
gpuNVIDIA H100 (sm_90)
Artifacts
No artifacts published with this run.
Replications and notes
No attestations yet.
Community attestations. They never change the evidence level; only a KernelIndex-controlled rerun does.
Add a reproduction or note
Canonical manifest
Show manifestHide manifest
{
"run": {
"kind": "BenchmarkRun",
"spec": {
"status": "passed",
"timing": {
"latencyNs": {
"median": 43490959
},
"primaryStatistic": "median"
},
"observedAt": "2024-10-02T16:21:37.000Z",
"measurements": [
{
"unit": "ns",
"value": 43474380,
"metric": "latency",
"statistic": "p20"
},
{
"unit": "ns",
"value": 43507542,
"metric": "latency",
"statistic": "p80"
}
],
"sourceNative": {
"source": "liger-kernel-bench",
"metrics": {
"ms_20": 43.47438049316406,
"ms_50": 43.49095916748047,
"ms_80": 43.50754165649414
},
"benchmark": "jsd/forward",
"externalId": "jsd/forward/liger/nvidia-h100-pcie/tokens8192-vocab65536-fp32/2024-10-02-16-21-37"
},
"protocolDigest": "sha256:a79eaf4df66c97c6adc1fb2d09df78682155d1ea86652ae0b70199875bde7747",
"workloadDigest": "sha256:0198db0958301e45a0537c064249129a7c3d1b651a85e78cdce2814e1f2f47c3",
"environmentDigest": "sha256:89f9b28b06758f22a1df331fea58bfde2c9cc0f3627d1829509743ea538f0011",
"implementationDigest": "sha256:cfb3edb701b526d49e58d1de743fa6b7ce8e9b2d82736c8a68b45577428f96fc"
},
"metadata": {
"name": "liger-jsd-forward-liger-nvidia-h100-pcie-tokens8192-vocab65536-fp32-2024-10-02-16-21-37",
"title": "Jensen-Shannon divergence loss · liger · forward",
"labels": {
"liger_version": "0.3.1"
}
},
"apiVersion": "kernelindex.dev/v1alpha1"
},
"protocol": {
"kind": "BenchmarkProtocol",
"spec": {
"harness": {
"name": "Liger-Kernel benchmark scripts",
"repository": "https://github.com/linkedin/Liger-Kernel"
},
"measurement": {
"timer": "cuda_events",
"primaryStatistic": "median"
},
"comparability": {
"notes": "Timed: the module forward pass, via triton.testing.do_bench with quantiles 0.5/0.2/0.8 (median with p20/p80 spread). The upstream CSV records no CUDA, driver, or torch version; the Liger release rides each run's labels. Comparable only within one kernel, workload, timed pass, and GPU.",
"family": "liger_kernel_bench"
}
},
"metadata": {
"name": "liger-bench-do-bench-forward",
"title": "Liger benchmark harness · forward pass"
},
"apiVersion": "kernelindex.dev/v1alpha1"
},
"environment": {
"kind": "ExecutionEnvironment",
"spec": {
"hardware": {
"vendor": "nvidia",
"product": "NVIDIA H100",
"formFactor": "PCIe",
"architecture": "sm_90"
},
"software": {}
},
"metadata": {
"name": "liger-bench-nvidia-h100-pcie",
"title": "Liger benchmark host · H100 PCIe"
},
"apiVersion": "kernelindex.dev/v1alpha1"
}
}Cite this record (permalink, digest, access date)
Report an issue with this run
Published 2026-08-18 · Liger-Kernel benchmarks · BSD-2-ClauseAll results for Jensen-Shannon divergence loss →JSON