[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6ADNKF7VS8XCWQ2TA1ZM92R":3,"skill-same:01M6ADNKF7VS8XCWQ2TA1ZM92R":108},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":18,"publishedAt":46,"linkOnly":47,"markdownContent":48,"markdownBody":49,"frontMatter":50,"refName":62,"commitSha":63,"skillDirectory":64,"forksCount":65,"repositoryHtmlUrl":66,"repositoryHomepageUrl":67,"ownerHtmlUrl":68,"ownerBlogUrl":69,"ownerDescription":70,"updatedAt":71,"install":72,"resources":85},"01M6ADNKF7VS8XCWQ2TA1ZM92R","distributed-training-megatron-core",null,"Trains large language models (2B-462B parameters) using NVIDIA Megatron-Core with advanced parallelism strategies. Use when training models >1B parameters, need maximum GPU efficiency (47% MFU on H100), or require tensor\u002Fpipeline\u002Fsequence\u002Fcontext\u002Fexpert parallelism. Production-ready framework used for Nemotron, LLaMA, DeepSeek.","davila7\u002Fclaude-code-templates","davila7","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F6216945?v=4","01M6ADEWWFPS191NFM0Y1EX1JH","Daniel Avila","01M6ADJHGM78PHQRD3XZTFDVKK",33481,"MIT","https:\u002F\u002Fgithub.com\u002Fdavila7\u002Fclaude-code-templates\u002Fblob\u002Fmain\u002Fcli-tool\u002Fcomponents\u002Fskills\u002Fai-research\u002Fdistributed-training-megatron-core\u002FSKILL.md",[],[19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45],"distributed","training","core","megatron core","Distributed Training","distributed training megatron core","megatron","model","when","production-ready","advanced","Model Parallelism","Megatron-Core","H100","apex","strategies","efficiency","expert","ready","large language models","pipeline","Large-Scale Training","Pipeline Parallelism","tensor","gpu","framework","research","2026-11-10 10:12:30",false,"---\nname: distributed-training-megatron-core\ndescription: Trains large language models (2B-462B parameters) using NVIDIA Megatron-Core with advanced parallelism strategies. Use when training models >1B parameters, need maximum GPU efficiency (47% MFU on H100), or require tensor\u002Fpipeline\u002Fsequence\u002Fcontext\u002Fexpert parallelism. Production-ready framework used for Nemotron, LLaMA, DeepSeek.\nlicense: MIT\nmetadata:\n  version: 1.0.0\n  author: Orchestra Research\n  tags: [Megatron-Core, Large-Scale Training, NVIDIA, Tensor Parallelism, Pipeline Parallelism, Model Parallelism, H100, Distributed Training, Production]\n  dependencies: [megatron-core, torch, apex, transformer-engine]\n---\n\n# Megatron-Core - Large-Scale LLM Training\n\n## Quick start\n\nMegatron-Core trains LLMs from 2B to 462B parameters with up to 47% Model FLOP Utilization on H100 GPUs through advanced parallelism strategies.\n\n**Installation**:\n```bash\n# Docker (recommended)\ndocker run --gpus all -it --rm nvcr.io\u002Fnvidia\u002Fpytorch:25.04-py3\n\n# Or pip\npip install megatron-core\n```\n\n**Simple distributed training**:\n```bash\n# Train with 2 GPUs using data parallelism\ntorchrun --nproc_per_node=2 examples\u002Frun_simple_mcore_train_loop.py\n\n# Or LLaMA-3 8B training\n.\u002Fexamples\u002Fllama\u002Ftrain_llama3_8b_fp8.sh\n```\n\n## Common workflows\n\n### Workflow 1: Train LLaMA-style model with 3D parallelism\n\nCopy this checklist:\n\n```\nLLaMA Training Setup:\n- [ ] Step 1: Choose parallelism configuration\n- [ ] Step 2: Configure training hyperparameters\n- [ ] Step 3: Launch distributed training\n- [ ] Step 4: Monitor performance metrics\n```\n\n**Step 1: Choose parallelism configuration**\n\nModel size determines parallelism strategy:\n\n| Model Size | GPUs | Tensor Parallel | Pipeline Parallel | Data Parallel | Context Parallel |\n|------------|------|-----------------|-------------------|---------------|------------------|\n| 7B | 8 | 1 | 1 | 8 | 1 |\n| 13B | 8 | 2 | 1 | 4 | 1 |\n| 70B | 64 | 4 | 4 | 4 | 1 |\n| 405B | 128 | 8 | 8 | 2 | 2 |\n\n**Step 2: Configure training hyperparameters**\n\n```bash\n#!\u002Fbin\u002Fbash\n# train_llama_70b.sh\n\nGPUS_PER_NODE=8\nNNODES=8  # 64 GPUs total\nTP=4      # Tensor parallel\nPP=4      # Pipeline parallel\nCP=1      # Context parallel\n\n# LLaMA 70B configuration\nMODEL_SIZE=70  # Billion parameters\nHIDDEN_SIZE=8192\nNUM_LAYERS=80\nNUM_HEADS=64\nSEQ_LENGTH=4096\n\n# Training hyperparameters\nMICRO_BATCH=1\nGLOBAL_BATCH=1024\nLR=3e-4\n\ntorchrun \\\n  --nproc_per_node=$GPUS_PER_NODE \\\n  --nnodes=$NNODES \\\n  pretrain_gpt.py \\\n  --tensor-model-parallel-size $TP \\\n  --pipeline-model-parallel-size $PP \\\n  --context-parallel-size $CP \\\n  --sequence-parallel \\\n  --num-layers $NUM_LAYERS \\\n  --hidden-size $HIDDEN_SIZE \\\n  --num-attention-heads $NUM_HEADS \\\n  --seq-length $SEQ_LENGTH \\\n  --max-position-embeddings $SEQ_LENGTH \\\n  --micro-batch-size $MICRO_BATCH \\\n  --global-batch-size $GLOBAL_BATCH \\\n  --lr $LR \\\n  --train-iters 100000 \\\n  --lr-decay-style cosine \\\n  --lr-warmup-iters 2000 \\\n  --weight-decay 0.1 \\\n  --clip-grad 1.0 \\\n  --bf16 \\\n  --use-mcore-models \\\n  --transformer-impl transformer_engine \\\n  --data-path \u002Fpath\u002Fto\u002Fdata \\\n  --vocab-file \u002Fpath\u002Fto\u002Fvocab.json \\\n  --merge-file \u002Fpath\u002Fto\u002Fmerges.txt\n```\n\n**Step 3: Launch distributed training**\n\n```bash\n# Single node (8 GPUs)\nbash train_llama_70b.sh\n\n# Multi-node with SLURM\nsbatch --nodes=8 --gpus-per-node=8 train_llama_70b.sh\n```\n\n**Step 4: Monitor performance metrics**\n\nKey metrics to track:\n```\nModel FLOP Utilization (MFU): Target >40% on H100\nThroughput: Tokens\u002Fsec\u002FGPU\nMemory usage: \u003C80GB per GPU for 70B model\nLoss: Should decrease steadily\n```\n\n### Workflow 2: Configure Mixture of Experts (MoE) training\n\nFor sparse MoE models like Mixtral.\n\n```\nMoE Training:\n- [ ] Step 1: Configure expert parallelism\n- [ ] Step 2: Set MoE hyperparameters\n- [ ] Step 3: Launch training with EP\n```\n\n**Step 1: Configure expert parallelism**\n\n```bash\n# Mixtral 8x7B example\nTENSOR_PARALLEL=2\nPIPELINE_PARALLEL=1\nEXPERT_PARALLEL=4  # Split 8 experts across 4 GPUs\nDATA_PARALLEL=4\n\nTOTAL_GPUS=$((TENSOR_PARALLEL * PIPELINE_PARALLEL * EXPERT_PARALLEL * DATA_PARALLEL))\n# = 2 * 1 * 4 * 4 = 32 GPUs\n```\n\n**Step 2: Set MoE hyperparameters**\n\n```bash\ntorchrun \\\n  --nproc_per_node=8 \\\n  pretrain_gpt.py \\\n  --tensor-model-parallel-size 2 \\\n  --pipeline-model-parallel-size 1 \\\n  --expert-model-parallel-size 4 \\\n  --num-experts 8 \\\n  --moe-router-topk 2 \\\n  --moe-router-load-balancing-type aux_loss \\\n  --moe-aux-loss-coeff 0.01 \\\n  --hidden-size 4096 \\\n  --num-layers 32 \\\n  --num-attention-heads 32 \\\n  --seq-length 4096 \\\n  --max-position-embeddings 4096 \\\n  --bf16 \\\n  --use-mcore-models \\\n  --transformer-impl transformer_engine \\\n  --data-path \u002Fpath\u002Fto\u002Fdata \\\n  --vocab-file \u002Fpath\u002Fto\u002Fvocab.json \\\n  --merge-file \u002Fpath\u002Fto\u002Fmerges.txt\n```\n\n**Step 3: Launch training with EP**\n\nExpert parallelism distributes different experts across GPUs, reducing memory while maintaining capacity.\n\n```\nMemory without EP: 8 experts × 7B = 56GB per GPU\nMemory with EP=4: 2 experts × 7B = 14GB per GPU\nSavings: 75% memory reduction\n```\n\n### Workflow 3: Optimize for maximum throughput\n\nAchieve 47% MFU on H100.\n\n```\nPerformance Optimization:\n- [ ] Step 1: Enable Flash Attention\n- [ ] Step 2: Use FP8 precision (H100)\n- [ ] Step 3: Optimize micro-batch size\n- [ ] Step 4: Tune parallelism degrees\n```\n\n**Step 1: Enable optimizations**\n\n```bash\n--use-mcore-models  # Use Megatron Core models\n--transformer-impl transformer_engine  # Use Transformer Engine\n--sequence-parallel  # Reduce activation memory (use with TP)\n```\n\n**Step 2: Use FP8 precision (H100 only)**\n\n```bash\n--fp8-hybrid  # FP8 mixed precision training\n# Transformer Engine handles FP8 automatically\n```\n\nResult: 1.5-2x speedup on H100 vs BF16.\n\n**Step 3: Optimize micro-batch size**\n\nFind largest micro-batch that fits in memory:\n\n```bash\n# Start with 1, increase until OOM\nfor MBS in 1 2 4 8; do\n  echo \"Testing micro-batch-size=$MBS\"\n  torchrun ... --micro-batch-size $MBS\ndone\n```\n\nTypical values:\n- 7B model: 4-8\n- 70B model: 1-2\n- 405B model: 1\n\n**Step 4: Tune parallelism degrees**\n\nRules of thumb:\n```\nTensor Parallel: Use ≤8 (limited by NVLink within node)\nPipeline Parallel: Use for >70B models\nContext Parallel: Use for sequences >8K tokens\nData Parallel: Fill remaining GPUs\n```\n\nExample 405B on 128 H100s:\n```\nTP=8 (1 node)\nPP=8 (across nodes)\nCP=2 (long sequences)\nDP=1\nTotal = 8 × 8 × 2 × 1 = 128 GPUs\n```\n\n## When to use vs alternatives\n\n**Use Megatron-Core when:**\n- Training models >10B parameters\n- Need maximum efficiency (target >40% MFU)\n- Using NVIDIA GPUs (A100, H100)\n- Production training at scale\n- Want fine-grained parallelism control\n\n**Use alternatives instead:**\n- **PyTorch FSDP**: Models \u003C70B, simpler API, PyTorch native\n- **DeepSpeed**: Easier setup, good for \u003C100B models\n- **HuggingFace Accelerate**: Prototyping, simpler workflows\n- **LitGPT**: Educational, single-file implementations\n\n## Common issues\n\n**Issue: Low GPU utilization (\u003C30% MFU)**\n\nCauses:\n1. Micro-batch too small\n2. Too much parallelism overhead\n3. Not using Flash Attention\n\nFixes:\n```bash\n# Increase micro-batch\n--micro-batch-size 4  # Was 1\n\n# Enable optimizations\n--use-flash-attn\n--sequence-parallel\n\n# Reduce TP if >8\n--tensor-model-parallel-size 4  # Was 16\n```\n\n**Issue: Out of memory**\n\nReduce memory with:\n```bash\n--tensor-model-parallel-size 2  # Split model across GPUs\n--recompute-granularity full  # Gradient checkpointing\n--recompute-method block  # Checkpoint transformer blocks\n--recompute-num-layers 1  # Checkpoint every layer\n```\n\nOr use CPU\u002FNVMe offloading:\n```bash\n--cpu-optimizer  # Offload optimizer to CPU\n--cpu-optimizer-type ADAM  # CPU Adam variant\n```\n\n**Issue: Training slower than expected**\n\nCheck:\n1. **Network bottleneck**: Ensure InfiniBand\u002FNVLink enabled\n2. **Pipeline bubbles**: Use interleaved pipeline schedule\n   ```bash\n   --num-layers-per-virtual-pipeline-stage 2\n   ```\n3. **Data loading**: Use fast data loader\n   ```bash\n   --dataloader-type cyclic\n   ```\n\n**Issue: Diverging loss**\n\nStabilize training:\n```bash\n--lr-warmup-iters 2000  # Longer warmup\n--clip-grad 1.0  # Gradient clipping\n--init-method-std 0.006  # Smaller init\n--attention-dropout 0.0  # No dropout in attention\n--hidden-dropout 0.0  # No dropout in FFN\n```\n\n## Advanced topics\n\n**Parallelism strategies**: See [references\u002Fparallelism-guide.md](references\u002Fparallelism-guide.md) for detailed comparison of TP\u002FPP\u002FDP\u002FCP\u002FEP with performance analysis and when to use each.\n\n**Performance benchmarks**: See [references\u002Fbenchmarks.md](references\u002Fbenchmarks.md) for MFU numbers across different model sizes and GPU configurations.\n\n**Production configurations**: See [references\u002Fproduction-examples.md](references\u002Fproduction-examples.md) for real-world setups from LLaMA 3 405B, Nemotron-4 340B, and DeepSeek-V3 671B.\n\n**Training recipes**: See [references\u002Ftraining-recipes.md](references\u002Ftraining-recipes.md) for complete hyperparameter configurations for GPT\u002FLLaMA\u002FMixtral architectures.\n\n## Hardware requirements\n\n- **GPU**: NVIDIA Ampere+ (A100, H100, B200)\n  - Turing works but slower\n  - FP8 requires Hopper\u002FAda\u002FBlackwell\n- **Network**: InfiniBand or 400Gb+ Ethernet for multi-node\n- **Memory per GPU**:\n  - 7B model: 40GB+\n  - 70B model: 80GB (with TP=4)\n  - 405B model: 80GB (with TP=8, PP=8)\n- **Storage**: Fast NVMe for checkpoints (1TB+ for 70B+ models)\n\n## Resources\n\n- Docs: https:\u002F\u002Fdocs.nvidia.com\u002Fmegatron-core\u002F\n- GitHub: https:\u002F\u002Fgithub.com\u002FNVIDIA\u002FMegatron-LM\n- Papers:\n  - \"Megatron-LM: Training Multi-Billion Parameter Language Models Using Model Parallelism\" (2019)\n  - \"Efficient Large-Scale Language Model Training on GPU Clusters Using Megatron-LM\" (2021)\n- NeMo Framework: https:\u002F\u002Fdocs.nvidia.com\u002Fnemo-framework\u002F (built on Megatron-Core)\n\n\n","# Megatron-Core - Large-Scale LLM Training\n\n## Quick start\n\nMegatron-Core trains LLMs from 2B to 462B parameters with up to 47% Model FLOP Utilization on H100 GPUs through advanced parallelism strategies.\n\n**Installation**:\n```bash\n# Docker (recommended)\ndocker run --gpus all -it --rm nvcr.io\u002Fnvidia\u002Fpytorch:25.04-py3\n\n# Or pip\npip install megatron-core\n```\n\n**Simple distributed training**:\n```bash\n# Train with 2 GPUs using data parallelism\ntorchrun --nproc_per_node=2 examples\u002Frun_simple_mcore_train_loop.py\n\n# Or LLaMA-3 8B training\n.\u002Fexamples\u002Fllama\u002Ftrain_llama3_8b_fp8.sh\n```\n\n## Common workflows\n\n### Workflow 1: Train LLaMA-style model with 3D parallelism\n\nCopy this checklist:\n\n```\nLLaMA Training Setup:\n- [ ] Step 1: Choose parallelism configuration\n- [ ] Step 2: Configure training hyperparameters\n- [ ] Step 3: Launch distributed training\n- [ ] Step 4: Monitor performance metrics\n```\n\n**Step 1: Choose parallelism configuration**\n\nModel size determines parallelism strategy:\n\n| Model Size | GPUs | Tensor Parallel | Pipeline Parallel | Data Parallel | Context Parallel |\n|------------|------|-----------------|-------------------|---------------|------------------|\n| 7B | 8 | 1 | 1 | 8 | 1 |\n| 13B | 8 | 2 | 1 | 4 | 1 |\n| 70B | 64 | 4 | 4 | 4 | 1 |\n| 405B | 128 | 8 | 8 | 2 | 2 |\n\n**Step 2: Configure training hyperparameters**\n\n```bash\n#!\u002Fbin\u002Fbash\n# train_llama_70b.sh\n\nGPUS_PER_NODE=8\nNNODES=8  # 64 GPUs total\nTP=4      # Tensor parallel\nPP=4      # Pipeline parallel\nCP=1      # Context parallel\n\n# LLaMA 70B configuration\nMODEL_SIZE=70  # Billion parameters\nHIDDEN_SIZE=8192\nNUM_LAYERS=80\nNUM_HEADS=64\nSEQ_LENGTH=4096\n\n# Training hyperparameters\nMICRO_BATCH=1\nGLOBAL_BATCH=1024\nLR=3e-4\n\ntorchrun \\\n  --nproc_per_node=$GPUS_PER_NODE \\\n  --nnodes=$NNODES \\\n  pretrain_gpt.py \\\n  --tensor-model-parallel-size $TP \\\n  --pipeline-model-parallel-size $PP \\\n  --context-parallel-size $CP \\\n  --sequence-parallel \\\n  --num-layers $NUM_LAYERS \\\n  --hidden-size $HIDDEN_SIZE \\\n  --num-attention-heads $NUM_HEADS \\\n  --seq-length $SEQ_LENGTH \\\n  --max-position-embeddings $SEQ_LENGTH \\\n  --micro-batch-size $MICRO_BATCH \\\n  --global-batch-size $GLOBAL_BATCH \\\n  --lr $LR \\\n  --train-iters 100000 \\\n  --lr-decay-style cosine \\\n  --lr-warmup-iters 2000 \\\n  --weight-decay 0.1 \\\n  --clip-grad 1.0 \\\n  --bf16 \\\n  --use-mcore-models \\\n  --transformer-impl transformer_engine \\\n  --data-path \u002Fpath\u002Fto\u002Fdata \\\n  --vocab-file \u002Fpath\u002Fto\u002Fvocab.json \\\n  --merge-file \u002Fpath\u002Fto\u002Fmerges.txt\n```\n\n**Step 3: Launch distributed training**\n\n```bash\n# Single node (8 GPUs)\nbash train_llama_70b.sh\n\n# Multi-node with SLURM\nsbatch --nodes=8 --gpus-per-node=8 train_llama_70b.sh\n```\n\n**Step 4: Monitor performance metrics**\n\nKey metrics to track:\n```\nModel FLOP Utilization (MFU): Target >40% on H100\nThroughput: Tokens\u002Fsec\u002FGPU\nMemory usage: \u003C80GB per GPU for 70B model\nLoss: Should decrease steadily\n```\n\n### Workflow 2: Configure Mixture of Experts (MoE) training\n\nFor sparse MoE models like Mixtral.\n\n```\nMoE Training:\n- [ ] Step 1: Configure expert parallelism\n- [ ] Step 2: Set MoE hyperparameters\n- [ ] Step 3: Launch training with EP\n```\n\n**Step 1: Configure expert parallelism**\n\n```bash\n# Mixtral 8x7B example\nTENSOR_PARALLEL=2\nPIPELINE_PARALLEL=1\nEXPERT_PARALLEL=4  # Split 8 experts across 4 GPUs\nDATA_PARALLEL=4\n\nTOTAL_GPUS=$((TENSOR_PARALLEL * PIPELINE_PARALLEL * EXPERT_PARALLEL * DATA_PARALLEL))\n# = 2 * 1 * 4 * 4 = 32 GPUs\n```\n\n**Step 2: Set MoE hyperparameters**\n\n```bash\ntorchrun \\\n  --nproc_per_node=8 \\\n  pretrain_gpt.py \\\n  --tensor-model-parallel-size 2 \\\n  --pipeline-model-parallel-size 1 \\\n  --expert-model-parallel-size 4 \\\n  --num-experts 8 \\\n  --moe-router-topk 2 \\\n  --moe-router-load-balancing-type aux_loss \\\n  --moe-aux-loss-coeff 0.01 \\\n  --hidden-size 4096 \\\n  --num-layers 32 \\\n  --num-attention-heads 32 \\\n  --seq-length 4096 \\\n  --max-position-embeddings 4096 \\\n  --bf16 \\\n  --use-mcore-models \\\n  --transformer-impl transformer_engine \\\n  --data-path \u002Fpath\u002Fto\u002Fdata \\\n  --vocab-file \u002Fpath\u002Fto\u002Fvocab.json \\\n  --merge-file \u002Fpath\u002Fto\u002Fmerges.txt\n```\n\n**Step 3: Launch training with EP**\n\nExpert parallelism distributes different experts across GPUs, reducing memory while maintaining capacity.\n\n```\nMemory without EP: 8 experts × 7B = 56GB per GPU\nMemory with EP=4: 2 experts × 7B = 14GB per GPU\nSavings: 75% memory reduction\n```\n\n### Workflow 3: Optimize for maximum throughput\n\nAchieve 47% MFU on H100.\n\n```\nPerformance Optimization:\n- [ ] Step 1: Enable Flash Attention\n- [ ] Step 2: Use FP8 precision (H100)\n- [ ] Step 3: Optimize micro-batch size\n- [ ] Step 4: Tune parallelism degrees\n```\n\n**Step 1: Enable optimizations**\n\n```bash\n--use-mcore-models  # Use Megatron Core models\n--transformer-impl transformer_engine  # Use Transformer Engine\n--sequence-parallel  # Reduce activation memory (use with TP)\n```\n\n**Step 2: Use FP8 precision (H100 only)**\n\n```bash\n--fp8-hybrid  # FP8 mixed precision training\n# Transformer Engine handles FP8 automatically\n```\n\nResult: 1.5-2x speedup on H100 vs BF16.\n\n**Step 3: Optimize micro-batch size**\n\nFind largest micro-batch that fits in memory:\n\n```bash\n# Start with 1, increase until OOM\nfor MBS in 1 2 4 8; do\n  echo \"Testing micro-batch-size=$MBS\"\n  torchrun ... --micro-batch-size $MBS\ndone\n```\n\nTypical values:\n- 7B model: 4-8\n- 70B model: 1-2\n- 405B model: 1\n\n**Step 4: Tune parallelism degrees**\n\nRules of thumb:\n```\nTensor Parallel: Use ≤8 (limited by NVLink within node)\nPipeline Parallel: Use for >70B models\nContext Parallel: Use for sequences >8K tokens\nData Parallel: Fill remaining GPUs\n```\n\nExample 405B on 128 H100s:\n```\nTP=8 (1 node)\nPP=8 (across nodes)\nCP=2 (long sequences)\nDP=1\nTotal = 8 × 8 × 2 × 1 = 128 GPUs\n```\n\n## When to use vs alternatives\n\n**Use Megatron-Core when:**\n- Training models >10B parameters\n- Need maximum efficiency (target >40% MFU)\n- Using NVIDIA GPUs (A100, H100)\n- Production training at scale\n- Want fine-grained parallelism control\n\n**Use alternatives instead:**\n- **PyTorch FSDP**: Models \u003C70B, simpler API, PyTorch native\n- **DeepSpeed**: Easier setup, good for \u003C100B models\n- **HuggingFace Accelerate**: Prototyping, simpler workflows\n- **LitGPT**: Educational, single-file implementations\n\n## Common issues\n\n**Issue: Low GPU utilization (\u003C30% MFU)**\n\nCauses:\n1. Micro-batch too small\n2. Too much parallelism overhead\n3. Not using Flash Attention\n\nFixes:\n```bash\n# Increase micro-batch\n--micro-batch-size 4  # Was 1\n\n# Enable optimizations\n--use-flash-attn\n--sequence-parallel\n\n# Reduce TP if >8\n--tensor-model-parallel-size 4  # Was 16\n```\n\n**Issue: Out of memory**\n\nReduce memory with:\n```bash\n--tensor-model-parallel-size 2  # Split model across GPUs\n--recompute-granularity full  # Gradient checkpointing\n--recompute-method block  # Checkpoint transformer blocks\n--recompute-num-layers 1  # Checkpoint every layer\n```\n\nOr use CPU\u002FNVMe offloading:\n```bash\n--cpu-optimizer  # Offload optimizer to CPU\n--cpu-optimizer-type ADAM  # CPU Adam variant\n```\n\n**Issue: Training slower than expected**\n\nCheck:\n1. **Network bottleneck**: Ensure InfiniBand\u002FNVLink enabled\n2. **Pipeline bubbles**: Use interleaved pipeline schedule\n   ```bash\n   --num-layers-per-virtual-pipeline-stage 2\n   ```\n3. **Data loading**: Use fast data loader\n   ```bash\n   --dataloader-type cyclic\n   ```\n\n**Issue: Diverging loss**\n\nStabilize training:\n```bash\n--lr-warmup-iters 2000  # Longer warmup\n--clip-grad 1.0  # Gradient clipping\n--init-method-std 0.006  # Smaller init\n--attention-dropout 0.0  # No dropout in attention\n--hidden-dropout 0.0  # No dropout in FFN\n```\n\n## Advanced topics\n\n**Parallelism strategies**: See [references\u002Fparallelism-guide.md](references\u002Fparallelism-guide.md) for detailed comparison of TP\u002FPP\u002FDP\u002FCP\u002FEP with performance analysis and when to use each.\n\n**Performance benchmarks**: See [references\u002Fbenchmarks.md](references\u002Fbenchmarks.md) for MFU numbers across different model sizes and GPU configurations.\n\n**Production configurations**: See [references\u002Fproduction-examples.md](references\u002Fproduction-examples.md) for real-world setups from LLaMA 3 405B, Nemotron-4 340B, and DeepSeek-V3 671B.\n\n**Training recipes**: See [references\u002Ftraining-recipes.md](references\u002Ftraining-recipes.md) for complete hyperparameter configurations for GPT\u002FLLaMA\u002FMixtral architectures.\n\n## Hardware requirements\n\n- **GPU**: NVIDIA Ampere+ (A100, H100, B200)\n  - Turing works but slower\n  - FP8 requires Hopper\u002FAda\u002FBlackwell\n- **Network**: InfiniBand or 400Gb+ Ethernet for multi-node\n- **Memory per GPU**:\n  - 7B model: 40GB+\n  - 70B model: 80GB (with TP=4)\n  - 405B model: 80GB (with TP=8, PP=8)\n- **Storage**: Fast NVMe for checkpoints (1TB+ for 70B+ models)\n\n## Resources\n\n- Docs: https:\u002F\u002Fdocs.nvidia.com\u002Fmegatron-core\u002F\n- GitHub: https:\u002F\u002Fgithub.com\u002FNVIDIA\u002FMegatron-LM\n- Papers:\n  - \"Megatron-LM: Training Multi-Billion Parameter Language Models Using Model Parallelism\" (2019)\n  - \"Efficient Large-Scale Language Model Training on GPU Clusters Using Megatron-LM\" (2021)\n- NeMo Framework: https:\u002F\u002Fdocs.nvidia.com\u002Fnemo-framework\u002F (built on Megatron-Core)\n\n\n",{"name":5,"license":15,"metadata":51,"description":7},{"tags":52,"author":56,"version":57,"dependencies":58},[31,40,53,54,41,30,32,23,55],"NVIDIA","Tensor Parallelism","Production","Orchestra Research","1.0.0",[59,60,33,61],"megatron-core","torch","transformer-engine","main","2b749c82b5d11fac39d9a6e021ce3f5b8a541bd1","cli-tool\u002Fcomponents\u002Fskills\u002Fai-research\u002Fdistributed-training-megatron-core",3931,"https:\u002F\u002Fgithub.com\u002Fdavila7\u002Fclaude-code-templates","https:\u002F\u002Faitmpl.com","https:\u002F\u002Fgithub.com\u002Fdavila7","https:\u002F\u002Fdanielavila.me","Building AI dev tools with LLMs","2026-11-14 06:14:46",{"source":8,"skillName":5,"commands":73,"prompt":83,"zipAvailable":84},[74,77,80],{"tool":75,"command":76},"npx","npx skills add davila7\u002Fclaude-code-templates --skill distributed-training-megatron-core",{"tool":78,"command":79},"bunx","bunx skills add davila7\u002Fclaude-code-templates --skill distributed-training-megatron-core",{"tool":81,"command":82},"pnpm","pnpm dlx skills add davila7\u002Fclaude-code-templates --skill distributed-training-megatron-core","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fdavila7\u002Fclaude-code-templates\u002Fblob\u002Fmain\u002Fcli-tool\u002Fcomponents\u002Fskills\u002Fai-research\u002Fdistributed-training-megatron-core\u002FSKILL.md",true,[86,92,96,100,104],{"path":87,"resourceType":88,"resourceTypeDesc":89,"sizeBytes":90,"sourceHtmlUrl":91},"agents\u002Fopenai.yaml","REFERENCE","参考文档",247,"https:\u002F\u002Fgithub.com\u002Fnodetool-ai\u002Fnodetool\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fimplement\u002Fagents\u002Fopenai.yaml",{"path":93,"resourceType":88,"resourceTypeDesc":89,"sizeBytes":94,"sourceHtmlUrl":95},"references\u002Fbenchmarks.md",7363,"https:\u002F\u002Fgithub.com\u002Fdavila7\u002Fclaude-code-templates\u002Fblob\u002Fmain\u002Fcli-tool\u002Fcomponents\u002Fskills\u002Fai-research\u002Fdistributed-training-megatron-core\u002Freferences\u002Fbenchmarks.md",{"path":97,"resourceType":88,"resourceTypeDesc":89,"sizeBytes":98,"sourceHtmlUrl":99},"references\u002Fparallelism-guide.md",9591,"https:\u002F\u002Fgithub.com\u002Fdavila7\u002Fclaude-code-templates\u002Fblob\u002Fmain\u002Fcli-tool\u002Fcomponents\u002Fskills\u002Fai-research\u002Fdistributed-training-megatron-core\u002Freferences\u002Fparallelism-guide.md",{"path":101,"resourceType":88,"resourceTypeDesc":89,"sizeBytes":102,"sourceHtmlUrl":103},"references\u002Fproduction-examples.md",10903,"https:\u002F\u002Fgithub.com\u002Fdavila7\u002Fclaude-code-templates\u002Fblob\u002Fmain\u002Fcli-tool\u002Fcomponents\u002Fskills\u002Fai-research\u002Fdistributed-training-megatron-core\u002Freferences\u002Fproduction-examples.md",{"path":105,"resourceType":88,"resourceTypeDesc":89,"sizeBytes":106,"sourceHtmlUrl":107},"references\u002Ftraining-recipes.md",11588,"https:\u002F\u002Fgithub.com\u002Fdavila7\u002Fclaude-code-templates\u002Fblob\u002Fmain\u002Fcli-tool\u002Fcomponents\u002Fskills\u002Fai-research\u002Fdistributed-training-megatron-core\u002Freferences\u002Ftraining-recipes.md",[109,140,167,200,235,259],{"skillNo":110,"name":111,"nameZh":6,"description":112,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":113,"categories":114,"keywords":120,"publishedAt":46},"01M70NHBTQ8BV0A2SCAZXEPHCX","wealthville-pool-scoring","Score DeFi liquidity pools before providing liquidity. Returns a 0-100 pool score and an Enter\u002FHold\u002FExit\u002FReduce\u002FAvoid verdict with calibrated confidence for ~68,800 Solana pools and 575 EVM pools, plus a public miss-inclusive track record. Use when asked whether a pool is worth LPing into, which pools look best right now, or how accurate the scores have actually been.","https:\u002F\u002Fgithub.com\u002Fdavila7\u002Fclaude-code-templates\u002Fblob\u002Fmain\u002Fcli-tool\u002Fcomponents\u002Fskills\u002Fweb-data\u002Fwealthville-pool-scoring\u002FSKILL.md",[115],{"categoryNo":116,"categoryName":117,"taxonomyType":118,"taxonomyTypeDesc":119},"01JBSEEDCAT000000000000023","Skill与智能体开发","SCENARIO","使用场景",[121,122,123,124,125,126,127,128,129,27,130,131,132,133,134,135,136,137,138,139],"scoring","pool","wealthville pool scoring","liquidity","which","for","exit","now","use","reduce","with","record","track","verdict","evm","solana","how","public","look",{"skillNo":141,"name":142,"nameZh":6,"description":143,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":144,"categories":145,"keywords":147,"publishedAt":46},"01M70MB6VR8ERJ4EM8D5KHBMEV","technical-debt-auditor","Scans a codebase and produces a quantified technical debt report with a prioritized remediation plan. Analyzes dead code, TODO\u002FFIXME\u002FHACK markers, dependency staleness, complexity hotspots, code duplication signals, test coverage gaps, unused exports, and outdated patterns. Use this skill when the user asks to audit their codebase for tech debt, wants a health check of their code quality, says things like \"where should I focus refactoring\", \"find code smells\", \"audit my codebase\", \"what technical debt do we have\", \"prioritize cleanup work\", or asks about code quality metrics. Also trigger when the user mentions \"code rot\", \"legacy cleanup\", \"refactoring priorities\", or wants to understand the overall health of a project before a major effort. Trigger even if the user doesn't say \"technical debt\" explicitly — any request to systematically assess code quality and identify improvement priorities belongs here.","https:\u002F\u002Fgithub.com\u002Fdavila7\u002Fclaude-code-templates\u002Fblob\u002Fmain\u002Fcli-tool\u002Fcomponents\u002Fskills\u002Fdevelopment\u002Ftechnical-debt-auditor\u002FSKILL.md",[146],{"categoryNo":116,"categoryName":117,"taxonomyType":118,"taxonomyTypeDesc":119},[148,149,150,151,152,153,154,155,156,157,158,27,159,160,161,162,163,164,165,166],"technical debt","debt","technical","technical debt auditor","auditor","coverage","complexity","markers","project","cleanup","unused","code","todo","work","plan","focus","signals","audit","dependency",{"skillNo":168,"name":169,"nameZh":6,"description":170,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":171,"categories":172,"keywords":182,"publishedAt":46},"01M70KPYH9QNHCNME5GJQGE22K","readonly-sql-guard","Make an agent's database access provably read-only. Use when an agent runs caller-supplied SQL against a production\u002FERP\u002Fanalytics database (a \"query\" MCP tool, text-to-SQL, an admin console), or when reviewing DB-access code for write\u002FDDL\u002Fside-effect\u002Ftransaction-escape risks. Checks what a statement CALLS, not just its shape.","https:\u002F\u002Fgithub.com\u002Fdavila7\u002Fclaude-code-templates\u002Fblob\u002Fmain\u002Fcli-tool\u002Fcomponents\u002Fskills\u002Fdatabase\u002Freadonly-sql-guard\u002FSKILL.md",[173,176,177],{"categoryNo":174,"categoryName":175,"taxonomyType":118,"taxonomyTypeDesc":119},"01JBSEEDCAT000000000000010","后端与接口",{"categoryNo":116,"categoryName":117,"taxonomyType":118,"taxonomyTypeDesc":119},{"categoryNo":178,"categoryName":179,"taxonomyType":180,"taxonomyTypeDesc":181},"01JBSEEDCAT000000000000033","数据分析师","PROFESSION","适用职业",[183,184,185,186,187,188,189,126,190,191,192,193,194,195,196,129,197,198,27,199],"readonly","guard","sql guard","sql","readonly sql guard","erp","effect","read-only","db","just","tool","calls","only","MCP","access","database","query",{"skillNo":201,"name":202,"nameZh":6,"description":203,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":204,"licenseFileUrl":205,"sourceHtmlUrl":206,"categories":207,"keywords":211,"publishedAt":46},"01M6AFY6F7CC2BBZP1QD6VTNVC","web-artifacts-builder","Suite of tools for creating elaborate, multi-component claude.ai HTML artifacts using modern frontend web technologies (React, Tailwind CSS, shadcn\u002Fui). Use for complex artifacts requiring state management, routing, or shadcn\u002Fui components - not for simple single-file HTML\u002FJSX artifacts.","Apache-2.0","https:\u002F\u002Fgithub.com\u002Fdavila7\u002Fclaude-code-templates\u002Fblob\u002Fmain\u002Fcli-tool\u002Fcomponents\u002Fskills\u002Futilities\u002Fweb-artifacts-builder\u002FLICENSE.txt","https:\u002F\u002Fgithub.com\u002Fdavila7\u002Fclaude-code-templates\u002Fblob\u002Fmain\u002Fcli-tool\u002Fcomponents\u002Fskills\u002Futilities\u002Fweb-artifacts-builder\u002FSKILL.md",[208],{"categoryNo":209,"categoryName":210,"taxonomyType":118,"taxonomyTypeDesc":119},"01JBSEEDCAT000000000000009","前端开发",[212,213,214,215,216,217,218,219,220,221,222,223,224,225,226,227,228,229,230,231,232,233,126,234],"artifacts builder","web","web artifacts builder","artifacts","builder","state","routing","state management","component","css","frontend","ui","Tools","file","components","shadcn","react","Tailwind","suite","jsx","html","single-file","ui components",{"skillNo":236,"name":237,"nameZh":6,"description":238,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":204,"licenseFileUrl":239,"sourceHtmlUrl":240,"categories":241,"keywords":246,"publishedAt":46},"01M6AFWF33SFV9Z3SF4PEEE4BF","webapp-testing","Toolkit for interacting with and testing local web applications using Playwright. Supports verifying frontend functionality, debugging UI behavior, capturing browser screenshots, and viewing browser logs.","https:\u002F\u002Fgithub.com\u002Fdavila7\u002Fclaude-code-templates\u002Fblob\u002Fmain\u002Fcli-tool\u002Fcomponents\u002Fskills\u002Fsecurity\u002Fwebapp-testing\u002FLICENSE.txt","https:\u002F\u002Fgithub.com\u002Fdavila7\u002Fclaude-code-templates\u002Fblob\u002Fmain\u002Fcli-tool\u002Fcomponents\u002Fskills\u002Fsecurity\u002Fwebapp-testing\u002FSKILL.md",[242,245],{"categoryNo":243,"categoryName":244,"taxonomyType":118,"taxonomyTypeDesc":119},"01JBSEEDCAT000000000000008","测试与质量",{"categoryNo":209,"categoryName":210,"taxonomyType":118,"taxonomyTypeDesc":119},[247,248,249,250,251,213,252,222,223,253,254,255,256,126,257,131,258],"webapp testing","testing","webapp","browser","playwright","applications","debugging","toolkit","logs","screenshots","local","behavior",{"skillNo":260,"name":261,"nameZh":6,"description":262,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":204,"licenseFileUrl":263,"sourceHtmlUrl":264,"categories":265,"keywords":273,"publishedAt":46},"01M6AF67Q67JJJ5V51JR2P52KA","discernment-nudge","After you give a substantive answer or draft that the user may act on — advice or recommendations, drafted artifacts such as goals, plans, pitches, proposals, or emails, estimates or projections, analysis or interpretation of data, factual claims they may rely on, or a multi-step argument — invoke this skill BEFORE finalizing your reply and then, if it applies, append 2-3 short follow-up questions, each tied to something specific in what you just produced, that help the user check key facts, probe the reasoning or assumptions, and notice missing context. Do this at most once per conversation. Skip it when the user asked a trivial how-to or simple lookup, wants a purely educational explanation, asked you only to format, convert, or assemble a file from content they provided, is writing code they will run, is doing creative writing or casual chat, or already asked you to double-check, cite, or review — the skill file explains these boundaries and the exact output format.","https:\u002F\u002Fgithub.com\u002Fdavila7\u002Fclaude-code-templates\u002Fblob\u002Fmain\u002Fcli-tool\u002Fcomponents\u002Fskills\u002Fproductivity\u002Fdiscernment-nudge\u002FLICENSE.txt","https:\u002F\u002Fgithub.com\u002Fdavila7\u002Fclaude-code-templates\u002Fblob\u002Fmain\u002Fcli-tool\u002Fcomponents\u002Fskills\u002Fproductivity\u002Fdiscernment-nudge\u002FSKILL.md",[266,269,270],{"categoryNo":267,"categoryName":268,"taxonomyType":118,"taxonomyTypeDesc":119},"01JBSEEDCAT000000000000018","写作与文案",{"categoryNo":116,"categoryName":117,"taxonomyType":118,"taxonomyTypeDesc":119},{"categoryNo":271,"categoryName":272,"taxonomyType":180,"taxonomyTypeDesc":181},"01JBSEEDCAT000000000000036","内容创作者",[274,275,276,277,278,279,280,281,215,282,225,159,283,284,285,286,287,195,288,27,289,290,291,292,293,137,294],"discernment nudge","nudge","interpretation","help","run","reasoning","data","writing","lookup","content","review","conversation","short","up","invoke","you","analysis","claims","argument","assemble","recommendations"]