Merge branch 'nv_main'

9df0c4a3 · wenjh · 0d874a4e · f122b07d · 9df0c4a3 · 9df0c4a3
Commit 9df0c4a3 authored Mar 03, 2026 by wenjh
20 changed files
--- a/docs/features/low_precision_training/performance_considerations/fused_layers_pytorch.py
+++ b/docs/features/low_precision_training/performance_considerations/fused_layers_pytorch.py
--- a/docs/features/low_precision_training/performance_considerations/img/fused_layers.svg
+++ b/docs/features/low_precision_training/performance_considerations/img/fused_layers.svg
--- a/docs/features/low_precision_training/performance_considerations/img/gemm_access_pattern.svg
+++ b/docs/features/low_precision_training/performance_considerations/img/gemm_access_pattern.svg
--- a/docs/features/low_precision_training/performance_considerations/img/hopper_vs_blackwell_layout.svg
+++ b/docs/features/low_precision_training/performance_considerations/img/hopper_vs_blackwell_layout.svg
--- a/docs/features/low_precision_training/performance_considerations/img/sequence_parallel_quantization.svg
+++ b/docs/features/low_precision_training/performance_considerations/img/sequence_parallel_quantization.svg
--- a/docs/features/low_precision_training/performance_considerations/img/transpose_fusion.svg
+++ b/docs/features/low_precision_training/performance_considerations/img/transpose_fusion.svg
--- a/docs/features/low_precision_training/performance_considerations/memory_usage_1_jax.out
+++ b/docs/features/low_precision_training/performance_considerations/memory_usage_1_jax.out
--- a/docs/features/low_precision_training/performance_considerations/memory_usage_1_jax.py
+++ b/docs/features/low_precision_training/performance_considerations/memory_usage_1_jax.py
--- a/docs/features/low_precision_training/performance_considerations/memory_usage_1_pytorch.out
+++ b/docs/features/low_precision_training/performance_considerations/memory_usage_1_pytorch.out
+# START_MEMORY_USAGE_1
+Memory usage after forward pass: 6.00 MB
+# END_MEMORY_USAGE_1
--- a/docs/features/low_precision_training/performance_considerations/memory_usage_1_pytorch.py
+++ b/docs/features/low_precision_training/performance_considerations/memory_usage_1_pytorch.py
--- a/docs/features/low_precision_training/performance_considerations/memory_usage_2_jax.out
+++ b/docs/features/low_precision_training/performance_considerations/memory_usage_2_jax.out
--- a/docs/features/low_precision_training/performance_considerations/memory_usage_2_jax.py
+++ b/docs/features/low_precision_training/performance_considerations/memory_usage_2_jax.py
--- a/docs/features/low_precision_training/performance_considerations/memory_usage_2_pytorch.out
+++ b/docs/features/low_precision_training/performance_considerations/memory_usage_2_pytorch.out
--- a/docs/features/low_precision_training/performance_considerations/memory_usage_2_pytorch.py
+++ b/docs/features/low_precision_training/performance_considerations/memory_usage_2_pytorch.py
--- a/docs/features/low_precision_training/performance_considerations/memory_usage_3_pytorch.out
+++ b/docs/features/low_precision_training/performance_considerations/memory_usage_3_pytorch.out
--- a/docs/features/low_precision_training/performance_considerations/memory_usage_3_pytorch.py
+++ b/docs/features/low_precision_training/performance_considerations/memory_usage_3_pytorch.py
--- a/docs/features/low_precision_training/performance_considerations/performance_considerations.rst
+++ b/docs/features/low_precision_training/performance_considerations/performance_considerations.rst
--- a/docs/features/low_precision_training/performance_considerations/save_original_input_pytorch.out
+++ b/docs/features/low_precision_training/performance_considerations/save_original_input_pytorch.out
--- a/docs/features/low_precision_training/performance_considerations/save_original_input_pytorch.py
+++ b/docs/features/low_precision_training/performance_considerations/save_original_input_pytorch.py
--- a/docs/index.rst
+++ b/docs/index.rst