mirror of
https://github.com/ROCm/ROCm.git
synced 2026-01-08 22:28:06 -05:00
Update training docs for Primus/25.11 (#5819)
* update conf and toc.yml.in * archive previous versions archive data files update anchors * primus pytorch: remove training batch size args * update primus megatron run cmds multi-node * update primus pytorch update * update update * update docker tag
This commit is contained in:
peterjunpark
@@ -163,6 +163,7 @@ article_pages = [
|
||||
{"file": "how-to/rocm-for-ai/training/benchmark-docker/previous-versions/jax-maxtext-v25.4", "os": ["linux"]},
|
||||
{"file": "how-to/rocm-for-ai/training/benchmark-docker/previous-versions/jax-maxtext-v25.5", "os": ["linux"]},
|
||||
{"file": "how-to/rocm-for-ai/training/benchmark-docker/mpt-llm-foundry", "os": ["linux"]},
|
||||
{"file": "how-to/rocm-for-ai/inference/xdit-diffusion-inference", "os": ["linux"]},
|
||||
|
||||
{"file": "how-to/rocm-for-ai/fine-tuning/index", "os": ["linux"]},
|
||||
{"file": "how-to/rocm-for-ai/fine-tuning/overview", "os": ["linux"]},
|
||||
|
||||
@@ -0,0 +1,49 @@
|
||||
docker:
|
||||
pull_tag: rocm/primus:v25.10
|
||||
docker_hub_url: https://hub.docker.com/layers/rocm/primus/v25.10/images/sha256-140c37cd2eeeb183759b9622543fc03cc210dc97cbfa18eeefdcbda84420c197
|
||||
components:
|
||||
ROCm: 7.1.0
|
||||
Primus: 0.3.0
|
||||
Primus Turbo: 0.1.1
|
||||
PyTorch: 2.10.0.dev20251112+rocm7.1
|
||||
Python: "3.10"
|
||||
Transformer Engine: 2.4.0.dev0+32e2d1d4
|
||||
Flash Attention: 2.8.3
|
||||
hipBLASLt: 1.2.0-09ab7153e2
|
||||
Triton: 3.4.0
|
||||
RCCL: 2.27.7
|
||||
model_groups:
|
||||
- group: Meta Llama
|
||||
tag: llama
|
||||
models:
|
||||
- model: Llama 3.3 70B
|
||||
mad_tag: pyt_megatron_lm_train_llama-3.3-70b
|
||||
- model: Llama 3.1 8B
|
||||
mad_tag: pyt_megatron_lm_train_llama-3.1-8b
|
||||
- model: Llama 3.1 70B
|
||||
mad_tag: pyt_megatron_lm_train_llama-3.1-70b
|
||||
- model: Llama 2 7B
|
||||
mad_tag: pyt_megatron_lm_train_llama-2-7b
|
||||
- model: Llama 2 70B
|
||||
mad_tag: pyt_megatron_lm_train_llama-2-70b
|
||||
- group: DeepSeek
|
||||
tag: deepseek
|
||||
models:
|
||||
- model: DeepSeek-V3 (proxy)
|
||||
mad_tag: pyt_megatron_lm_train_deepseek-v3-proxy
|
||||
- model: DeepSeek-V2-Lite
|
||||
mad_tag: pyt_megatron_lm_train_deepseek-v2-lite-16b
|
||||
- group: Mistral AI
|
||||
tag: mistral
|
||||
models:
|
||||
- model: Mixtral 8x7B
|
||||
mad_tag: pyt_megatron_lm_train_mixtral-8x7b
|
||||
- model: Mixtral 8x22B (proxy)
|
||||
mad_tag: pyt_megatron_lm_train_mixtral-8x22b-proxy
|
||||
- group: Qwen
|
||||
tag: qwen
|
||||
models:
|
||||
- model: Qwen 2.5 7B
|
||||
mad_tag: pyt_megatron_lm_train_qwen2.5-7b
|
||||
- model: Qwen 2.5 72B
|
||||
mad_tag: pyt_megatron_lm_train_qwen2.5-72b
|
||||
@@ -0,0 +1,58 @@
|
||||
docker:
|
||||
pull_tag: rocm/primus:v25.10
|
||||
docker_hub_url: https://hub.docker.com/layers/rocm/primus/v25.10/images/sha256-140c37cd2eeeb183759b9622543fc03cc210dc97cbfa18eeefdcbda84420c197
|
||||
components:
|
||||
ROCm: 7.1.0
|
||||
PyTorch: 2.10.0.dev20251112+rocm7.1
|
||||
Python: "3.10"
|
||||
Transformer Engine: 2.4.0.dev0+32e2d1d4
|
||||
Flash Attention: 2.8.3
|
||||
hipBLASLt: 1.2.0-09ab7153e2
|
||||
Triton: 3.4.0
|
||||
RCCL: 2.27.7
|
||||
model_groups:
|
||||
- group: Meta Llama
|
||||
tag: llama
|
||||
models:
|
||||
- model: Llama 3.3 70B
|
||||
mad_tag: primus_pyt_megatron_lm_train_llama-3.3-70b
|
||||
config_name: llama3.3_70B-pretrain.yaml
|
||||
- model: Llama 3.1 70B
|
||||
mad_tag: primus_pyt_megatron_lm_train_llama-3.1-70b
|
||||
config_name: llama3.1_70B-pretrain.yaml
|
||||
- model: Llama 3.1 8B
|
||||
mad_tag: primus_pyt_megatron_lm_train_llama-3.1-8b
|
||||
config_name: llama3.1_8B-pretrain.yaml
|
||||
- model: Llama 2 7B
|
||||
mad_tag: primus_pyt_megatron_lm_train_llama-2-7b
|
||||
config_name: llama2_7B-pretrain.yaml
|
||||
- model: Llama 2 70B
|
||||
mad_tag: primus_pyt_megatron_lm_train_llama-2-70b
|
||||
config_name: llama2_70B-pretrain.yaml
|
||||
- group: DeepSeek
|
||||
tag: deepseek
|
||||
models:
|
||||
- model: DeepSeek-V3 (proxy)
|
||||
mad_tag: primus_pyt_megatron_lm_train_deepseek-v3-proxy
|
||||
config_name: deepseek_v3-pretrain.yaml
|
||||
- model: DeepSeek-V2-Lite
|
||||
mad_tag: primus_pyt_megatron_lm_train_deepseek-v2-lite-16b
|
||||
config_name: deepseek_v2_lite-pretrain.yaml
|
||||
- group: Mistral AI
|
||||
tag: mistral
|
||||
models:
|
||||
- model: Mixtral 8x7B
|
||||
mad_tag: primus_pyt_megatron_lm_train_mixtral-8x7b
|
||||
config_name: mixtral_8x7B_v0.1-pretrain.yaml
|
||||
- model: Mixtral 8x22B (proxy)
|
||||
mad_tag: primus_pyt_megatron_lm_train_mixtral-8x22b-proxy
|
||||
config_name: mixtral_8x22B_v0.1-pretrain.yaml
|
||||
- group: Qwen
|
||||
tag: qwen
|
||||
models:
|
||||
- model: Qwen 2.5 7B
|
||||
mad_tag: primus_pyt_megatron_lm_train_qwen2.5-7b
|
||||
config_name: primus_qwen2.5_7B-pretrain.yaml
|
||||
- model: Qwen 2.5 72B
|
||||
mad_tag: primus_pyt_megatron_lm_train_qwen2.5-72b
|
||||
config_name: qwen2.5_72B-pretrain.yaml
|
||||
@@ -0,0 +1,32 @@
|
||||
docker:
|
||||
pull_tag: rocm/primus:v25.10
|
||||
docker_hub_url: https://hub.docker.com/layers/rocm/primus/v25.10/images/sha256-140c37cd2eeeb183759b9622543fc03cc210dc97cbfa18eeefdcbda84420c197
|
||||
components:
|
||||
ROCm: 7.1.0
|
||||
PyTorch: 2.10.0.dev20251112+rocm7.1
|
||||
Python: "3.10"
|
||||
Transformer Engine: 2.4.0.dev0+32e2d1d4
|
||||
Flash Attention: 2.8.3
|
||||
hipBLASLt: 1.2.0-09ab7153e2
|
||||
model_groups:
|
||||
- group: Meta Llama
|
||||
tag: llama
|
||||
models:
|
||||
- model: Llama 3.1 8B
|
||||
mad_tag: primus_pyt_train_llama-3.1-8b
|
||||
model_repo: Llama-3.1-8B
|
||||
url: https://huggingface.co/meta-llama/Llama-3.1-8B
|
||||
precision: BF16
|
||||
- model: Llama 3.1 70B
|
||||
mad_tag: primus_pyt_train_llama-3.1-70b
|
||||
model_repo: Llama-3.1-70B
|
||||
url: https://huggingface.co/meta-llama/Llama-3.1-70B
|
||||
precision: BF16
|
||||
- group: DeepSeek
|
||||
tag: deepseek
|
||||
models:
|
||||
- model: DeepSeek V2 16B
|
||||
mad_tag: primus_pyt_train_deepseek-v2
|
||||
model_repo: DeepSeek-V2
|
||||
url: https://huggingface.co/deepseek-ai/DeepSeek-V2
|
||||
precision: BF16
|
||||
@@ -0,0 +1,197 @@
|
||||
docker:
|
||||
pull_tag: rocm/primus:v25.10
|
||||
docker_hub_url: https://hub.docker.com/layers/rocm/primus/v25.10/images/sha256-140c37cd2eeeb183759b9622543fc03cc210dc97cbfa18eeefdcbda84420c197
|
||||
components:
|
||||
ROCm: 7.1.0
|
||||
Primus: 0.3.0
|
||||
Primus Turbo: 0.1.1
|
||||
PyTorch: 2.10.0.dev20251112+rocm7.1
|
||||
Python: "3.10"
|
||||
Transformer Engine: 2.4.0.dev0+32e2d1d4
|
||||
Flash Attention: 2.8.3
|
||||
hipBLASLt: 1.2.0-09ab7153e2
|
||||
model_groups:
|
||||
- group: Meta Llama
|
||||
tag: llama
|
||||
models:
|
||||
- model: Llama 4 Scout 17B-16E
|
||||
mad_tag: pyt_train_llama-4-scout-17b-16e
|
||||
model_repo: Llama-4-17B_16E
|
||||
url: https://huggingface.co/meta-llama/Llama-4-Scout-17B-16E
|
||||
precision: BF16
|
||||
training_modes: [finetune_fw, finetune_lora]
|
||||
- model: Llama 3.3 70B
|
||||
mad_tag: pyt_train_llama-3.3-70b
|
||||
model_repo: Llama-3.3-70B
|
||||
url: https://huggingface.co/meta-llama/Llama-3.3-70B-Instruct
|
||||
precision: BF16
|
||||
training_modes: [finetune_fw, finetune_lora, finetune_qlora]
|
||||
- model: Llama 3.2 1B
|
||||
mad_tag: pyt_train_llama-3.2-1b
|
||||
model_repo: Llama-3.2-1B
|
||||
url: https://huggingface.co/meta-llama/Llama-3.2-1B
|
||||
precision: BF16
|
||||
training_modes: [finetune_fw, finetune_lora]
|
||||
- model: Llama 3.2 3B
|
||||
mad_tag: pyt_train_llama-3.2-3b
|
||||
model_repo: Llama-3.2-3B
|
||||
url: https://huggingface.co/meta-llama/Llama-3.2-3B
|
||||
precision: BF16
|
||||
training_modes: [finetune_fw, finetune_lora]
|
||||
- model: Llama 3.2 Vision 11B
|
||||
mad_tag: pyt_train_llama-3.2-vision-11b
|
||||
model_repo: Llama-3.2-Vision-11B
|
||||
url: https://huggingface.co/meta-llama/Llama-3.2-11B-Vision
|
||||
precision: BF16
|
||||
training_modes: [finetune_fw]
|
||||
- model: Llama 3.2 Vision 90B
|
||||
mad_tag: pyt_train_llama-3.2-vision-90b
|
||||
model_repo: Llama-3.2-Vision-90B
|
||||
url: https://huggingface.co/meta-llama/Llama-3.2-90B-Vision
|
||||
precision: BF16
|
||||
training_modes: [finetune_fw]
|
||||
- model: Llama 3.1 8B
|
||||
mad_tag: pyt_train_llama-3.1-8b
|
||||
model_repo: Llama-3.1-8B
|
||||
url: https://huggingface.co/meta-llama/Llama-3.1-8B
|
||||
precision: BF16
|
||||
training_modes: [pretrain, finetune_fw, finetune_lora, HF_pretrain]
|
||||
- model: Llama 3.1 70B
|
||||
mad_tag: pyt_train_llama-3.1-70b
|
||||
model_repo: Llama-3.1-70B
|
||||
url: https://huggingface.co/meta-llama/Llama-3.1-70B-Instruct
|
||||
precision: BF16
|
||||
training_modes: [pretrain, finetune_fw, finetune_lora]
|
||||
- model: Llama 3.1 405B
|
||||
mad_tag: pyt_train_llama-3.1-405b
|
||||
model_repo: Llama-3.1-405B
|
||||
url: https://huggingface.co/meta-llama/Llama-3.1-405B
|
||||
precision: BF16
|
||||
training_modes: [finetune_qlora]
|
||||
- model: Llama 3 8B
|
||||
mad_tag: pyt_train_llama-3-8b
|
||||
model_repo: Llama-3-8B
|
||||
url: https://huggingface.co/meta-llama/Meta-Llama-3-8B
|
||||
precision: BF16
|
||||
training_modes: [finetune_fw, finetune_lora]
|
||||
- model: Llama 3 70B
|
||||
mad_tag: pyt_train_llama-3-70b
|
||||
model_repo: Llama-3-70B
|
||||
url: https://huggingface.co/meta-llama/Meta-Llama-3-70B
|
||||
precision: BF16
|
||||
training_modes: [finetune_fw, finetune_lora]
|
||||
- model: Llama 2 7B
|
||||
mad_tag: pyt_train_llama-2-7b
|
||||
model_repo: Llama-2-7B
|
||||
url: https://github.com/meta-llama/llama-models/tree/main/models/llama2
|
||||
precision: BF16
|
||||
training_modes: [finetune_fw, finetune_lora, finetune_qlora]
|
||||
- model: Llama 2 13B
|
||||
mad_tag: pyt_train_llama-2-13b
|
||||
model_repo: Llama-2-13B
|
||||
url: https://github.com/meta-llama/llama-models/tree/main/models/llama2
|
||||
precision: BF16
|
||||
training_modes: [finetune_fw, finetune_lora]
|
||||
- model: Llama 2 70B
|
||||
mad_tag: pyt_train_llama-2-70b
|
||||
model_repo: Llama-2-70B
|
||||
url: https://github.com/meta-llama/llama-models/tree/main/models/llama2
|
||||
precision: BF16
|
||||
training_modes: [finetune_lora, finetune_qlora]
|
||||
- group: OpenAI
|
||||
tag: openai
|
||||
models:
|
||||
- model: GPT OSS 20B
|
||||
mad_tag: pyt_train_gpt_oss_20b
|
||||
model_repo: GPT-OSS-20B
|
||||
url: https://huggingface.co/openai/gpt-oss-20b
|
||||
precision: BF16
|
||||
training_modes: [HF_finetune_lora]
|
||||
- model: GPT OSS 120B
|
||||
mad_tag: pyt_train_gpt_oss_120b
|
||||
model_repo: GPT-OSS-120B
|
||||
url: https://huggingface.co/openai/gpt-oss-120b
|
||||
precision: BF16
|
||||
training_modes: [HF_finetune_lora]
|
||||
- group: DeepSeek
|
||||
tag: deepseek
|
||||
models:
|
||||
- model: DeepSeek V2 16B
|
||||
mad_tag: primus_pyt_train_deepseek-v2
|
||||
model_repo: DeepSeek-V2
|
||||
url: https://huggingface.co/deepseek-ai/DeepSeek-V2
|
||||
precision: BF16
|
||||
training_modes: [pretrain]
|
||||
- group: Qwen
|
||||
tag: qwen
|
||||
models:
|
||||
- model: Qwen 3 8B
|
||||
mad_tag: pyt_train_qwen3-8b
|
||||
model_repo: Qwen3-8B
|
||||
url: https://huggingface.co/Qwen/Qwen3-8B
|
||||
precision: BF16
|
||||
training_modes: [finetune_fw, finetune_lora]
|
||||
- model: Qwen 3 32B
|
||||
mad_tag: pyt_train_qwen3-32b
|
||||
model_repo: Qwen3-32
|
||||
url: https://huggingface.co/Qwen/Qwen3-32B
|
||||
precision: BF16
|
||||
training_modes: [finetune_lora]
|
||||
- model: Qwen 2.5 32B
|
||||
mad_tag: pyt_train_qwen2.5-32b
|
||||
model_repo: Qwen2.5-32B
|
||||
url: https://huggingface.co/Qwen/Qwen2.5-32B
|
||||
precision: BF16
|
||||
training_modes: [finetune_lora]
|
||||
- model: Qwen 2.5 72B
|
||||
mad_tag: pyt_train_qwen2.5-72b
|
||||
model_repo: Qwen2.5-72B
|
||||
url: https://huggingface.co/Qwen/Qwen2.5-72B
|
||||
precision: BF16
|
||||
training_modes: [finetune_lora]
|
||||
- model: Qwen 2 1.5B
|
||||
mad_tag: pyt_train_qwen2-1.5b
|
||||
model_repo: Qwen2-1.5B
|
||||
url: https://huggingface.co/Qwen/Qwen2-1.5B
|
||||
precision: BF16
|
||||
training_modes: [finetune_fw, finetune_lora]
|
||||
- model: Qwen 2 7B
|
||||
mad_tag: pyt_train_qwen2-7b
|
||||
model_repo: Qwen2-7B
|
||||
url: https://huggingface.co/Qwen/Qwen2-7B
|
||||
precision: BF16
|
||||
training_modes: [finetune_fw, finetune_lora]
|
||||
- group: Stable Diffusion
|
||||
tag: sd
|
||||
models:
|
||||
- model: Stable Diffusion XL
|
||||
mad_tag: pyt_huggingface_stable_diffusion_xl_2k_lora_finetuning
|
||||
model_repo: SDXL
|
||||
url: https://huggingface.co/stabilityai/stable-diffusion-xl-base-1.0
|
||||
precision: BF16
|
||||
training_modes: [posttrain]
|
||||
- group: Flux
|
||||
tag: flux
|
||||
models:
|
||||
- model: FLUX.1-dev
|
||||
mad_tag: pyt_train_flux
|
||||
model_repo: Flux
|
||||
url: https://huggingface.co/black-forest-labs/FLUX.1-dev
|
||||
precision: BF16
|
||||
training_modes: [posttrain]
|
||||
- group: NCF
|
||||
tag: ncf
|
||||
models:
|
||||
- model: NCF
|
||||
mad_tag: pyt_ncf_training
|
||||
model_repo:
|
||||
url: https://github.com/ROCm/FluxBenchmark
|
||||
precision: FP32
|
||||
- group: DLRM
|
||||
tag: dlrm
|
||||
models:
|
||||
- model: DLRM v2
|
||||
mad_tag: pyt_train_dlrm
|
||||
model_repo: DLRM
|
||||
url: https://github.com/AMD-AGI/DLRMBenchmark
|
||||
training_modes: [pretrain]
|
||||
@@ -1,5 +1,5 @@
|
||||
docker:
|
||||
pull_tag: rocm/primus:v25.10
|
||||
pull_tag: rocm/primus:v25.11
|
||||
docker_hub_url: https://hub.docker.com/layers/rocm/primus/v25.10/images/sha256-140c37cd2eeeb183759b9622543fc03cc210dc97cbfa18eeefdcbda84420c197
|
||||
components:
|
||||
ROCm: 7.1.0
|
||||
|
||||
@@ -1,5 +1,5 @@
|
||||
docker:
|
||||
pull_tag: rocm/primus:v25.10
|
||||
pull_tag: rocm/primus:v25.11
|
||||
docker_hub_url: https://hub.docker.com/layers/rocm/primus/v25.10/images/sha256-140c37cd2eeeb183759b9622543fc03cc210dc97cbfa18eeefdcbda84420c197
|
||||
components:
|
||||
ROCm: 7.1.0
|
||||
@@ -25,8 +25,8 @@ model_groups:
|
||||
- group: DeepSeek
|
||||
tag: deepseek
|
||||
models:
|
||||
- model: DeepSeek V2 16B
|
||||
mad_tag: primus_pyt_train_deepseek-v2
|
||||
model_repo: DeepSeek-V2
|
||||
url: https://huggingface.co/deepseek-ai/DeepSeek-V2
|
||||
- model: DeepSeek V3 16B
|
||||
mad_tag: primus_pyt_train_deepseek-v3-16b
|
||||
model_repo: DeepSeek-V3
|
||||
url: https://huggingface.co/deepseek-ai/DeepSeek-V3
|
||||
precision: BF16
|
||||
|
||||
@@ -52,7 +52,7 @@ accelerate training workloads:
|
||||
- {{ component_version }}
|
||||
{% endfor %}
|
||||
|
||||
.. _amd-megatron-lm-model-support-v2510:
|
||||
.. _amd-megatron-lm-model-support-v25.11:
|
||||
|
||||
Supported models
|
||||
================
|
||||
@@ -97,7 +97,7 @@ accelerate training workloads:
|
||||
Some models, such as Llama, require an external license agreement through
|
||||
a third party (for example, Meta).
|
||||
|
||||
.. _amd-megatron-lm-performance-measurements-v2510:
|
||||
.. _amd-megatron-lm-performance-measurements-v25.11:
|
||||
|
||||
Performance measurements
|
||||
========================
|
||||
@@ -129,7 +129,7 @@ To test for optimal performance, consult the recommended :ref:`System health ben
|
||||
<rocm-for-ai-system-health-bench>`. This suite of tests will help you verify and fine-tune your
|
||||
system's configuration.
|
||||
|
||||
.. _mi300x-amd-megatron-lm-training-v2510:
|
||||
.. _mi300x-amd-megatron-lm-training-v25.11:
|
||||
|
||||
Environment setup
|
||||
=================
|
||||
@@ -138,7 +138,7 @@ Use the following instructions to set up the environment, configure the script t
|
||||
reproduce the benchmark results on MI300X Series GPUs with the AMD Megatron-LM Docker
|
||||
image.
|
||||
|
||||
.. _amd-megatron-lm-requirements-v2510:
|
||||
.. _amd-megatron-lm-requirements-v25.11:
|
||||
|
||||
Download the Docker image
|
||||
-------------------------
|
||||
@@ -190,7 +190,7 @@ Download the Docker image
|
||||
The Docker container hosts a verified commit of
|
||||
`<https://github.com/ROCm/Megatron-LM/tree/rocm_dev>`__.
|
||||
|
||||
.. _amd-megatron-lm-environment-setup-v2510:
|
||||
.. _amd-megatron-lm-environment-setup-v25.11:
|
||||
|
||||
Configuration
|
||||
=============
|
||||
@@ -200,39 +200,39 @@ Configuration
|
||||
Update the ``train_llama3.sh`` configuration script in the ``examples/llama``
|
||||
directory of
|
||||
`<https://github.com/ROCm/Megatron-LM/tree/rocm_dev/examples/llama>`__ to configure your training run.
|
||||
Options can also be passed as command line arguments as described in :ref:`Run training <amd-megatron-lm-run-training-v2510>`.
|
||||
Options can also be passed as command line arguments as described in :ref:`Run training <amd-megatron-lm-run-training-v25.11>`.
|
||||
|
||||
.. container:: model-doc pyt_megatron_lm_train_llama-2-7b pyt_megatron_lm_train_llama-2-70b
|
||||
|
||||
Update the ``train_llama2.sh`` configuration script in the ``examples/llama``
|
||||
directory of
|
||||
`<https://github.com/ROCm/Megatron-LM/tree/rocm_dev/examples/llama>`__ to configure your training run.
|
||||
Options can also be passed as command line arguments as described in :ref:`Run training <amd-megatron-lm-run-training-v2510>`.
|
||||
Options can also be passed as command line arguments as described in :ref:`Run training <amd-megatron-lm-run-training-v25.11>`.
|
||||
|
||||
.. container:: model-doc pyt_megatron_lm_train_deepseek-v3-proxy
|
||||
|
||||
Update the ``train_deepseekv3.sh`` configuration script in the ``examples/deepseek_v3``
|
||||
directory of
|
||||
`<https://github.com/ROCm/Megatron-LM/tree/rocm_dev/examples/deepseek_v3>`__ to configure your training run.
|
||||
Options can also be passed as command line arguments as described in :ref:`Run training <amd-megatron-lm-run-training-v2510>`.
|
||||
Options can also be passed as command line arguments as described in :ref:`Run training <amd-megatron-lm-run-training-v25.11>`.
|
||||
|
||||
.. container:: model-doc pyt_megatron_lm_train_deepseek-v2-lite-16b
|
||||
|
||||
Update the ``train_deepseekv2.sh`` configuration script in the ``examples/deepseek_v2``
|
||||
directory of
|
||||
`<https://github.com/ROCm/Megatron-LM/tree/rocm_dev/examples/deepseek_v2>`__ to configure your training run.
|
||||
Options can also be passed as command line arguments as described in :ref:`Run training <amd-megatron-lm-run-training-v2510>`.
|
||||
Options can also be passed as command line arguments as described in :ref:`Run training <amd-megatron-lm-run-training-v25.11>`.
|
||||
|
||||
.. container:: model-doc pyt_megatron_lm_train_mixtral-8x7b pyt_megatron_lm_train_mixtral-8x22b-proxy
|
||||
|
||||
Update the ``train_mixtral_moe.sh`` configuration script in the ``examples/mixtral``
|
||||
directory of
|
||||
`<https://github.com/ROCm/Megatron-LM/tree/rocm_dev/examples/mixtral>`__ to configure your training run.
|
||||
Options can also be passed as command line arguments as described in :ref:`Run training <amd-megatron-lm-run-training-v2510>`.
|
||||
Options can also be passed as command line arguments as described in :ref:`Run training <amd-megatron-lm-run-training-v25.11>`.
|
||||
|
||||
.. note::
|
||||
|
||||
See :ref:`Key options <amd-megatron-lm-benchmark-test-vars-v2510>` for more information on configuration options.
|
||||
See :ref:`Key options <amd-megatron-lm-benchmark-test-vars-v25.11>` for more information on configuration options.
|
||||
|
||||
Multi-node configuration
|
||||
------------------------
|
||||
@@ -240,7 +240,7 @@ Multi-node configuration
|
||||
Refer to :doc:`/how-to/rocm-for-ai/system-setup/multi-node-setup` to configure your environment for multi-node
|
||||
training. See :ref:`amd-megatron-lm-multi-node-examples` for example run commands.
|
||||
|
||||
.. _amd-megatron-lm-tokenizer-v2510:
|
||||
.. _amd-megatron-lm-tokenizer-v25.11:
|
||||
|
||||
Tokenizer
|
||||
---------
|
||||
@@ -377,7 +377,7 @@ Download the dataset
|
||||
|
||||
``TOKENIZER_MODEL`` can be any accessible Hugging Face tokenizer.
|
||||
Remember to either pre-download the tokenizer or setup Hugging Face access
|
||||
otherwise when needed -- see the :ref:`Tokenizer <amd-megatron-lm-tokenizer-v2510>` section.
|
||||
otherwise when needed -- see the :ref:`Tokenizer <amd-megatron-lm-tokenizer-v25.11>` section.
|
||||
|
||||
.. note::
|
||||
|
||||
@@ -479,13 +479,13 @@ Download the dataset
|
||||
|
||||
Ensure that the files are accessible inside the Docker container.
|
||||
|
||||
.. _amd-megatron-lm-run-training-v2510:
|
||||
.. _amd-megatron-lm-run-training-v25.11:
|
||||
|
||||
Run training
|
||||
============
|
||||
|
||||
Use the following example commands to set up the environment, configure
|
||||
:ref:`key options <amd-megatron-lm-benchmark-test-vars-v2510>`, and run training on
|
||||
:ref:`key options <amd-megatron-lm-benchmark-test-vars-v25.11>`, and run training on
|
||||
MI300X Series GPUs with the AMD Megatron-LM environment.
|
||||
|
||||
Before starting training, export the following environment variables.
|
||||
@@ -920,7 +920,7 @@ Single node training
|
||||
RECOMPUTE_ACTIVATIONS=full \
|
||||
CKPT_FORMAT=torch_dist
|
||||
|
||||
.. _amd-megatron-lm-multi-node-examples-v2510:
|
||||
.. _amd-megatron-lm-multi-node-examples-v25.11:
|
||||
|
||||
Multi-node training examples
|
||||
----------------------------
|
||||
@@ -971,7 +971,7 @@ training on 16 nodes, try the following command:
|
||||
|
||||
sbatch examples/deepseek_v3/train_deepseek_v3_slurm.sh
|
||||
|
||||
.. _amd-megatron-lm-benchmark-test-vars-v2510:
|
||||
.. _amd-megatron-lm-benchmark-test-vars-v25.11:
|
||||
|
||||
Key options
|
||||
-----------
|
||||
|
||||
@@ -16,7 +16,7 @@ previous releases of the ``ROCm/megatron-lm`` Docker image on `Docker Hub <https
|
||||
- Components
|
||||
- Resources
|
||||
|
||||
* - v25.10 (latest)
|
||||
* - v25.11
|
||||
-
|
||||
* ROCm 7.1.0
|
||||
* PyTorch 2.10.0.dev20251112+rocm7.1
|
||||
@@ -25,6 +25,15 @@ previous releases of the ``ROCm/megatron-lm`` Docker image on `Docker Hub <https
|
||||
* :doc:`Megatron-LM (legacy) documentation <../megatron-lm>`
|
||||
* `Docker Hub <https://hub.docker.com/layers/rocm/primus/v25.10/images/sha256-140c37cd2eeeb183759b9622543fc03cc210dc97cbfa18eeefdcbda84420c197>`__
|
||||
|
||||
* - v25.10
|
||||
-
|
||||
* ROCm 7.1.0
|
||||
* PyTorch 2.10.0.dev20251112+rocm7.1
|
||||
-
|
||||
* :doc:`Primus Megatron documentation <primus-megatron-v25.10>`
|
||||
* :doc:`Megatron-LM (legacy) documentation <megatron-lm-v25.10>`
|
||||
* `Docker Hub <https://hub.docker.com/layers/rocm/primus/v25.10/images/sha256-140c37cd2eeeb183759b9622543fc03cc210dc97cbfa18eeefdcbda84420c197>`__
|
||||
|
||||
* - v25.9
|
||||
-
|
||||
* ROCm 7.0.0
|
||||
|
||||
File diff suppressed because it is too large
Load Diff
File diff suppressed because it is too large
Load Diff
@@ -0,0 +1,448 @@
|
||||
:orphan:
|
||||
|
||||
.. meta::
|
||||
:description: How to train a model using PyTorch for ROCm.
|
||||
:keywords: ROCm, AI, LLM, train, PyTorch, torch, Llama, flux, tutorial, docker
|
||||
|
||||
****************************************
|
||||
Training a model with Primus and PyTorch
|
||||
****************************************
|
||||
|
||||
.. caution::
|
||||
|
||||
This documentation does not reflect the latest version of ROCm Primus PyTorch training
|
||||
performance benchmark documentation. See :doc:`../primus-pytorch` for the latest version.
|
||||
|
||||
`Primus <https://github.com/AMD-AGI/Primus>`__ is a unified and flexible
|
||||
LLM training framework designed to streamline training. It streamlines LLM
|
||||
training on AMD Instinct GPUs using a modular, reproducible configuration paradigm.
|
||||
Primus now supports the PyTorch torchtitan backend.
|
||||
|
||||
.. note::
|
||||
|
||||
For a unified training solution on AMD GPUs with ROCm, the `rocm/pytorch-training
|
||||
<https://hub.docker.com/r/rocm/pytorch-training/>`__ Docker Hub registry will be
|
||||
deprecated soon in favor of `rocm/primus <https://hub.docker.com/r/rocm/primus>`__.
|
||||
The ``rocm/primus`` Docker containers will cover PyTorch training ecosystem frameworks,
|
||||
including torchtitan and :doc:`Megatron-LM <primus-megatron>`.
|
||||
|
||||
Primus with the PyTorch torchtitan backend is designed to replace the
|
||||
:doc:`ROCm PyTorch training <pytorch-training>` workflow. See
|
||||
:doc:`pytorch-training` to see steps to run workloads without Primus.
|
||||
|
||||
AMD provides a ready-to-use Docker image for MI355X, MI350X, MI325X, and
|
||||
MI300X GPUs containing essential components for Primus and PyTorch training
|
||||
with Primus Turbo optimizations.
|
||||
|
||||
.. datatemplate:yaml:: /data/how-to/rocm-for-ai/training/primus-pytorch-benchmark-models.yaml
|
||||
|
||||
.. tab-set::
|
||||
|
||||
.. tab-item:: {{ data.docker.pull_tag }}
|
||||
:sync: {{ data.docker.pull_tag }}
|
||||
|
||||
.. list-table::
|
||||
:header-rows: 1
|
||||
|
||||
* - Software component
|
||||
- Version
|
||||
|
||||
{% for component_name, component_version in data.docker.components.items() %}
|
||||
* - {{ component_name }}
|
||||
- {{ component_version }}
|
||||
{% endfor %}
|
||||
|
||||
.. _amd-primus-pytorch-model-support-v2510:
|
||||
|
||||
Supported models
|
||||
================
|
||||
|
||||
The following models are pre-optimized for performance on the AMD Instinct MI325X and MI300X GPUs.
|
||||
Some instructions, commands, and training recommendations in this documentation might
|
||||
vary by model -- select one to get started.
|
||||
|
||||
.. datatemplate:yaml:: /data/how-to/rocm-for-ai/training/primus-pytorch-benchmark-models.yaml
|
||||
|
||||
{% set model_groups = data.model_groups %}
|
||||
.. raw:: html
|
||||
|
||||
<div id="vllm-benchmark-ud-params-picker" class="container-fluid">
|
||||
<div class="row gx-0">
|
||||
<div class="col-2 me-1 px-2 model-param-head">Model</div>
|
||||
<div class="row col-10 pe-0">
|
||||
{% for model_group in model_groups %}
|
||||
<div class="col-6 px-2 model-param" data-param-k="model-group" data-param-v="{{ model_group.tag }}" tabindex="0">{{ model_group.group }}</div>
|
||||
{% endfor %}
|
||||
</div>
|
||||
</div>
|
||||
|
||||
<div class="row gx-0 pt-1">
|
||||
<div class="col-2 me-1 px-2 model-param-head">Variant</div>
|
||||
<div class="row col-10 pe-0">
|
||||
{% for model_group in model_groups %}
|
||||
{% set models = model_group.models %}
|
||||
{% for model in models %}
|
||||
{% if models|length % 3 == 0 %}
|
||||
<div class="col-4 px-2 model-param" data-param-k="model" data-param-v="{{ model.mad_tag }}" data-param-group="{{ model_group.tag }}" tabindex="0">{{ model.model }}</div>
|
||||
{% else %}
|
||||
<div class="col-6 px-2 model-param" data-param-k="model" data-param-v="{{ model.mad_tag }}" data-param-group="{{ model_group.tag }}" tabindex="0">{{ model.model }}</div>
|
||||
{% endif %}
|
||||
{% endfor %}
|
||||
{% endfor %}
|
||||
</div>
|
||||
</div>
|
||||
</div>
|
||||
|
||||
.. seealso::
|
||||
|
||||
For additional workloads, including Llama 3.3, Llama 3.2, Llama 2, GPT OSS, Qwen, and Flux models,
|
||||
see the documentation :doc:`pytorch-training` (without Primus)
|
||||
|
||||
.. _amd-primus-pytorch-performance-measurements-v2510:
|
||||
|
||||
System validation
|
||||
=================
|
||||
|
||||
Before running AI workloads, it's important to validate that your AMD hardware is configured
|
||||
correctly and performing optimally.
|
||||
|
||||
If you have already validated your system settings, including aspects like NUMA auto-balancing, you
|
||||
can skip this step. Otherwise, complete the procedures in the :ref:`System validation and
|
||||
optimization <rocm-for-ai-system-optimization>` guide to properly configure your system settings
|
||||
before starting training.
|
||||
|
||||
To test for optimal performance, consult the recommended :ref:`System health benchmarks
|
||||
<rocm-for-ai-system-health-bench>`. This suite of tests will help you verify and fine-tune your
|
||||
system's configuration.
|
||||
|
||||
This Docker image is optimized for specific model configurations outlined
|
||||
below. Performance can vary for other training workloads, as AMD
|
||||
doesn’t test configurations and run conditions outside those described.
|
||||
|
||||
Pull the Docker image
|
||||
=====================
|
||||
|
||||
.. datatemplate:yaml:: /data/how-to/rocm-for-ai/training/primus-pytorch-benchmark-models.yaml
|
||||
|
||||
Use the following command to pull the Docker image from Docker Hub.
|
||||
|
||||
.. code-block:: shell
|
||||
|
||||
docker pull {{ data.docker.pull_tag }}
|
||||
|
||||
Run training
|
||||
============
|
||||
|
||||
Once the setup is complete, choose between the following two workflows to start benchmarking training.
|
||||
For fine-tuning workloads and multi-node training examples, see :doc:`pytorch-training` (without Primus).
|
||||
For best performance on MI325X, MI350X, and MI355X GPUs, you might need to
|
||||
tweak some configurations (such as batch sizes).
|
||||
|
||||
.. datatemplate:yaml:: /data/how-to/rocm-for-ai/training/primus-pytorch-benchmark-models.yaml
|
||||
|
||||
{% set docker = data.docker %}
|
||||
{% set model_groups = data.model_groups %}
|
||||
|
||||
.. tab-set::
|
||||
|
||||
.. tab-item:: MAD-integrated benchmarking
|
||||
|
||||
{% for model_group in model_groups %}
|
||||
{% for model in model_group.models %}
|
||||
|
||||
.. container:: model-doc {{ model.mad_tag }}
|
||||
|
||||
The following run command is tailored to {{ model.model }}.
|
||||
See :ref:`amd-primus-pytorch-model-support-v2510` to switch to another available model.
|
||||
|
||||
1. Clone the ROCm Model Automation and Dashboarding (`<https://github.com/ROCm/MAD>`__) repository to a local
|
||||
directory and install the required packages on the host machine.
|
||||
|
||||
.. code-block:: shell
|
||||
|
||||
git clone https://github.com/ROCm/MAD
|
||||
cd MAD
|
||||
pip install -r requirements.txt
|
||||
|
||||
2. For example, use this command to run the performance benchmark test on the {{ model.model }} model
|
||||
using one node with the {{ model.precision }} data type on the host machine.
|
||||
|
||||
.. code-block:: shell
|
||||
|
||||
export MAD_SECRETS_HFTOKEN="your personal Hugging Face token to access gated models"
|
||||
madengine run \
|
||||
--tags {{ model.mad_tag }} \
|
||||
--keep-model-dir \
|
||||
--live-output \
|
||||
--timeout 28800
|
||||
|
||||
MAD launches a Docker container with the name
|
||||
``container_ci-{{ model.mad_tag }}``. The latency and throughput reports of the
|
||||
model are collected in ``~/MAD/perf.csv``.
|
||||
|
||||
{% endfor %}
|
||||
{% endfor %}
|
||||
|
||||
.. tab-item:: Primus benchmarking
|
||||
|
||||
{% for model_group in model_groups %}
|
||||
{% for model in model_group.models %}
|
||||
|
||||
.. container:: model-doc {{ model.mad_tag }}
|
||||
|
||||
The following run commands are tailored to {{ model.model }}.
|
||||
See :ref:`amd-primus-pytorch-model-support-v2510` to switch to another available model.
|
||||
|
||||
.. rubric:: Download the Docker image and required packages
|
||||
|
||||
1. Pull the ``{{ docker.pull_tag }}`` Docker image from Docker Hub.
|
||||
|
||||
.. code-block:: shell
|
||||
|
||||
docker pull {{ docker.pull_tag }}
|
||||
|
||||
2. Run the Docker container.
|
||||
|
||||
.. code-block:: shell
|
||||
|
||||
docker run -it \
|
||||
--device /dev/dri \
|
||||
--device /dev/kfd \
|
||||
--network host \
|
||||
--ipc host \
|
||||
--group-add video \
|
||||
--cap-add SYS_PTRACE \
|
||||
--security-opt seccomp=unconfined \
|
||||
--privileged \
|
||||
-v $HOME:$HOME \
|
||||
-v $HOME/.ssh:/root/.ssh \
|
||||
--shm-size 64G \
|
||||
--name training_env \
|
||||
{{ docker.pull_tag }}
|
||||
|
||||
Use these commands if you exit the ``training_env`` container and need to return to it.
|
||||
|
||||
.. code-block:: shell
|
||||
|
||||
docker start training_env
|
||||
docker exec -it training_env bash
|
||||
|
||||
.. rubric:: Prepare training datasets and dependencies
|
||||
|
||||
The following benchmarking examples require downloading models and datasets
|
||||
from Hugging Face. To ensure successful access to gated repos, set your
|
||||
``HF_TOKEN``.
|
||||
|
||||
.. code-block:: shell
|
||||
|
||||
export HF_TOKEN=$your_personal_hugging_face_access_token
|
||||
|
||||
.. rubric:: Pretraining
|
||||
|
||||
To get started, navigate to the ``Primus`` directory in your container.
|
||||
|
||||
.. code-block::
|
||||
|
||||
cd /workspace/Primus
|
||||
|
||||
Now, to start the pretraining benchmark, use the ``run_pretrain.sh`` script
|
||||
included with Primus with the appropriate options.
|
||||
|
||||
.. rubric:: Benchmarking examples
|
||||
|
||||
.. container:: model-doc primus_pyt_train_llama-3.1-8b
|
||||
|
||||
Use the following command to run train Llama 3.1 8B with BF16 precision using Primus torchtitan.
|
||||
|
||||
.. tab-set::
|
||||
|
||||
.. tab-item:: MI355X and MI350X
|
||||
:sync: MI355X
|
||||
|
||||
.. code-block:: shell
|
||||
|
||||
EXP=examples/torchtitan/configs/MI355X/llama3.1_8B-BF16-pretrain.yaml \
|
||||
bash examples/run_pretrain.sh --training.local_batch_size 6
|
||||
|
||||
.. tab-item:: MI325X
|
||||
:sync: MI325X
|
||||
|
||||
.. code-block:: shell
|
||||
|
||||
EXP=examples/torchtitan/configs/MI300X/llama3.1_8B-BF16-pretrain.yaml \
|
||||
bash examples/run_pretrain.sh --training.local_batch_size 6
|
||||
|
||||
.. tab-item:: MI300X
|
||||
:sync: MI300X
|
||||
|
||||
.. code-block:: shell
|
||||
|
||||
EXP=examples/torchtitan/configs/MI300X/llama3.1_8B-BF16-pretrain.yaml \
|
||||
bash examples/run_pretrain.sh --training.local_batch_size 4
|
||||
|
||||
|
||||
To train Llama 3.1 8B with FP8 precision, use the following command.
|
||||
|
||||
.. tab-set::
|
||||
|
||||
.. tab-item:: MI355X and MI350X
|
||||
:sync: MI355X
|
||||
|
||||
.. code-block:: shell
|
||||
|
||||
EXP=examples/torchtitan/configs/MI355X/llama3.1_8B-BF16-pretrain.yaml \
|
||||
bash examples/run_pretrain.sh --training.local_batch_size 8
|
||||
|
||||
.. tab-item:: MI325X
|
||||
:sync: MI325X
|
||||
|
||||
.. code-block:: shell
|
||||
|
||||
EXP=examples/torchtitan/configs/MI300X/llama3.1_8B-FP8-pretrain.yaml \
|
||||
bash examples/run_pretrain.sh --training.local_batch_size 7
|
||||
|
||||
.. tab-item:: MI300X
|
||||
:sync: MI300X
|
||||
|
||||
.. code-block:: shell
|
||||
|
||||
EXP=examples/torchtitan/configs/MI300X/llama3.1_8B-FP8-pretrain.yaml \
|
||||
bash examples/run_pretrain.sh --training.local_batch_size 5
|
||||
|
||||
.. container:: model-doc primus_pyt_train_llama-3.1-70b
|
||||
|
||||
Use the following command to run train Llama 3.1 70B with BF16 precision using Primus torchtitan.
|
||||
|
||||
.. tab-set::
|
||||
|
||||
.. tab-item:: MI355X and MI350X
|
||||
:sync: MI355X and MI300X
|
||||
|
||||
.. code-block:: shell
|
||||
|
||||
EXP=examples/torchtitan/configs/MI355X/llama3.1_70B-BF16-pretrain.yaml \
|
||||
bash examples/run_pretrain.sh --training.local_batch_size 8
|
||||
|
||||
.. tab-item:: MI325X
|
||||
:sync: MI325X
|
||||
|
||||
.. code-block:: shell
|
||||
|
||||
EXP=examples/torchtitan/configs/MI300X/llama3.1_70B-BF16-pretrain.yaml \
|
||||
bash examples/run_pretrain.sh --training.local_batch_size 6
|
||||
|
||||
.. tab-item:: MI300X
|
||||
:sync: MI300X
|
||||
|
||||
.. code-block:: shell
|
||||
|
||||
EXP=examples/torchtitan/configs/MI300X/llama3.1_70B-BF16-pretrain.yaml \
|
||||
bash examples/run_pretrain.sh --training.local_batch_size 4
|
||||
|
||||
To train Llama 3.1 70B with FP8 precision, use the following command.
|
||||
|
||||
.. tab-set::
|
||||
|
||||
.. tab-item:: MI355X and MI350X
|
||||
:sync: MI355X
|
||||
|
||||
.. code-block:: shell
|
||||
|
||||
EXP=examples/torchtitan/configs/MI355X/llama3.1_70B-FP8-pretrain.yaml \
|
||||
bash examples/run_pretrain.sh --training.local_batch_size 6
|
||||
|
||||
.. tab-item:: MI325X
|
||||
:sync: MI325X
|
||||
|
||||
.. code-block:: shell
|
||||
|
||||
EXP=examples/torchtitan/configs/MI300X/llama3.1_70B-FP8-pretrain.yaml \
|
||||
bash examples/run_pretrain.sh --training.local_batch_size 5
|
||||
|
||||
.. tab-item:: MI300X
|
||||
:sync: MI300X
|
||||
|
||||
.. code-block:: shell
|
||||
|
||||
EXP=examples/torchtitan/configs/MI300X/llama3.1_70B-FP8-pretrain.yaml \
|
||||
bash examples/run_pretrain.sh --training.local_batch_size 3
|
||||
|
||||
.. container:: model-doc primus_pyt_train_deepseek-v2
|
||||
|
||||
Use the following command to run train DeepSeek V2 16B with BF16 precision using Primus torchtitan.
|
||||
|
||||
.. tab-set::
|
||||
|
||||
.. tab-item:: MI355X and MI350X
|
||||
:sync: MI355X and MI300X
|
||||
|
||||
.. code-block:: shell
|
||||
|
||||
EXP=examples/torchtitan/configs/MI355X/deepseek_v3_16b-pretrain.yaml \
|
||||
bash examples/run_pretrain.sh --training.local_batch_size 16
|
||||
|
||||
.. tab-item:: MI325X
|
||||
:sync: MI325X
|
||||
|
||||
.. code-block:: shell
|
||||
|
||||
EXP=examples/torchtitan/configs/MI300X/deepseek_v3_16b-pretrain.yaml \
|
||||
bash examples/run_pretrain.sh --training.local_batch_size 10
|
||||
|
||||
.. tab-item:: MI300X
|
||||
:sync: MI300X
|
||||
|
||||
.. code-block:: shell
|
||||
|
||||
EXP=examples/torchtitan/configs/MI300X/deepseek_v3_16b-pretrain.yaml \
|
||||
bash examples/run_pretrain.sh --training.local_batch_size 8
|
||||
|
||||
To train DeepSeek V2 16B with FP8 precision, use the following command.
|
||||
|
||||
.. tab-set::
|
||||
|
||||
.. tab-item:: MI355X and MI350X
|
||||
:sync: MI355X
|
||||
|
||||
.. code-block:: shell
|
||||
|
||||
EXP=examples/torchtitan/configs/MI355X/deepseek_v3_16b-pretrain.yaml \
|
||||
bash examples/run_pretrain.sh --training.local_batch_size 16
|
||||
|
||||
.. tab-item:: MI325X
|
||||
:sync: MI325X
|
||||
|
||||
.. code-block:: shell
|
||||
|
||||
EXP=examples/torchtitan/configs/MI300X/deepseek_v3_16b-pretrain.yaml \
|
||||
bash examples/run_pretrain.sh --training.local_batch_size 8
|
||||
|
||||
.. tab-item:: MI300X
|
||||
:sync: MI300X
|
||||
|
||||
.. code-block:: shell
|
||||
|
||||
EXP=examples/torchtitan/configs/MI300X/deepseek_v3_16b-pretrain.yaml \
|
||||
bash examples/run_pretrain.sh --training.local_batch_size 8
|
||||
{% endfor %}
|
||||
{% endfor %}
|
||||
|
||||
Further reading
|
||||
===============
|
||||
|
||||
- For an introduction to Primus, see `Primus: A Lightweight, Unified Training
|
||||
Framework for Large Models on AMD GPUs <https://rocm.blogs.amd.com/software-tools-optimization/primus/README.html>`__.
|
||||
|
||||
- To learn more about MAD and the ``madengine`` CLI, see the `MAD usage guide <https://github.com/ROCm/MAD?tab=readme-ov-file#usage-guide>`__.
|
||||
|
||||
- To learn more about system settings and management practices to configure your system for
|
||||
AMD Instinct MI300X Series GPUs, see `AMD Instinct MI300X system optimization <https://instinct.docs.amd.com/projects/amdgpu-docs/en/latest/system-optimization/mi300x.html>`_.
|
||||
|
||||
- For a list of other ready-made Docker images for AI with ROCm, see
|
||||
`AMD Infinity Hub <https://www.amd.com/en/developer/resources/infinity-hub.html#f-amd_hub_category=AI%20%26%20ML%20Models>`_.
|
||||
|
||||
Previous versions
|
||||
=================
|
||||
|
||||
See :doc:`pytorch-training-history` to find documentation for previous releases
|
||||
of the ``ROCm/pytorch-training`` Docker image.
|
||||
@@ -16,7 +16,7 @@ previous releases of the ``ROCm/pytorch-training`` Docker image on `Docker Hub <
|
||||
- Components
|
||||
- Resources
|
||||
|
||||
* - v25.10 (latest)
|
||||
* - v25.11
|
||||
-
|
||||
* ROCm 7.1.0
|
||||
* PyTorch 2.10.0.dev20251112+rocm7.1
|
||||
@@ -25,6 +25,15 @@ previous releases of the ``ROCm/pytorch-training`` Docker image on `Docker Hub <
|
||||
* :doc:`PyTorch training (legacy) documentation <../pytorch-training>`
|
||||
* `Docker Hub <https://hub.docker.com/layers/rocm/primus/v25.10/images/sha256-140c37cd2eeeb183759b9622543fc03cc210dc97cbfa18eeefdcbda84420c197>`__
|
||||
|
||||
* - v25.10
|
||||
-
|
||||
* ROCm 7.1.0
|
||||
* PyTorch 2.10.0.dev20251112+rocm7.1
|
||||
-
|
||||
* :doc:`Primus PyTorch Training documentation <primus-pytorch-v25.10>`
|
||||
* :doc:`PyTorch training (legacy) documentation <pytorch-training-v25.10>`
|
||||
* `Docker Hub <https://hub.docker.com/layers/rocm/primus/v25.10/images/sha256-140c37cd2eeeb183759b9622543fc03cc210dc97cbfa18eeefdcbda84420c197>`__
|
||||
|
||||
* - v25.9
|
||||
-
|
||||
* ROCm 7.0.0
|
||||
|
||||
@@ -0,0 +1,669 @@
|
||||
:orphan:
|
||||
|
||||
.. meta::
|
||||
:description: How to train a model using PyTorch for ROCm.
|
||||
:keywords: ROCm, AI, LLM, train, PyTorch, torch, Llama, flux, tutorial, docker
|
||||
|
||||
**************************************
|
||||
Training a model with PyTorch on ROCm
|
||||
**************************************
|
||||
|
||||
.. caution::
|
||||
|
||||
This documentation does not reflect the latest version of ROCm PyTorch training
|
||||
performance benchmark documentation. See :doc:`../pytorch-training` for the latest version.
|
||||
|
||||
.. note::
|
||||
|
||||
For a unified training solution on AMD GPUs with ROCm, the `rocm/pytorch-training
|
||||
<https://hub.docker.com/r/rocm/pytorch-training/>`__ Docker Hub registry will be
|
||||
deprecated soon in favor of `rocm/primus <https://hub.docker.com/r/rocm/primus>`__.
|
||||
The ``rocm/primus`` Docker containers will cover PyTorch training ecosystem frameworks,
|
||||
including torchtitan and :doc:`Megatron-LM <../primus-megatron>`.
|
||||
|
||||
See :doc:`../primus-pytorch` for details.
|
||||
|
||||
PyTorch is an open-source machine learning framework that is widely used for
|
||||
model training with GPU-optimized components for transformer-based models.
|
||||
The PyTorch for ROCm training Docker image provides a prebuilt optimized
|
||||
environment for fine-tuning and pretraining a model on AMD Instinct MI325X
|
||||
and MI300X GPUs. It includes the following software components to accelerate
|
||||
training workloads:
|
||||
|
||||
.. datatemplate:yaml:: /data/how-to/rocm-for-ai/training/pytorch-training-benchmark-models.yaml
|
||||
|
||||
.. tab-set::
|
||||
|
||||
.. tab-item:: {{ data.docker.pull_tag }}
|
||||
:sync: {{ data.docker.pull_tag }}
|
||||
|
||||
.. list-table::
|
||||
:header-rows: 1
|
||||
|
||||
* - Software component
|
||||
- Version
|
||||
|
||||
{% for component_name, component_version in data.docker.components.items() %}
|
||||
* - {{ component_name }}
|
||||
- {{ component_version }}
|
||||
{% endfor %}
|
||||
|
||||
.. _amd-pytorch-training-model-support-v2510:
|
||||
|
||||
Supported models
|
||||
================
|
||||
|
||||
The following models are pre-optimized for performance on the AMD Instinct
|
||||
MI355X, MI350X, MI325X, and MI300X GPUs. Some instructions, commands, and
|
||||
training recommendations in this documentation might vary by model -- select
|
||||
one to get started.
|
||||
|
||||
.. datatemplate:yaml:: /data/how-to/rocm-for-ai/training/pytorch-training-benchmark-models.yaml
|
||||
|
||||
{% set model_groups = data.model_groups %}
|
||||
.. raw:: html
|
||||
|
||||
<div id="vllm-benchmark-ud-params-picker" class="container-fluid">
|
||||
<div class="row gx-0">
|
||||
<div class="col-2 me-1 px-2 model-param-head">Model</div>
|
||||
<div class="row col-10 pe-0">
|
||||
{% for model_group in model_groups %}
|
||||
<div class="col-4 px-2 model-param" data-param-k="model-group" data-param-v="{{ model_group.tag }}" tabindex="0">{{ model_group.group }}</div>
|
||||
{% endfor %}
|
||||
</div>
|
||||
</div>
|
||||
|
||||
<div class="row gx-0 pt-1">
|
||||
<div class="col-2 me-1 px-2 model-param-head">Variant</div>
|
||||
<div class="row col-10 pe-0">
|
||||
{% for model_group in model_groups %}
|
||||
{% set models = model_group.models %}
|
||||
{% for model in models %}
|
||||
{% if models|length % 3 == 0 %}
|
||||
<div class="col-4 px-2 model-param" data-param-k="model" data-param-v="{{ model.mad_tag }}" data-param-group="{{ model_group.tag }}" tabindex="0">{{ model.model }}</div>
|
||||
{% else %}
|
||||
<div class="col-6 px-2 model-param" data-param-k="model" data-param-v="{{ model.mad_tag }}" data-param-group="{{ model_group.tag }}" tabindex="0">{{ model.model }}</div>
|
||||
{% endif %}
|
||||
{% endfor %}
|
||||
{% endfor %}
|
||||
</div>
|
||||
</div>
|
||||
</div>
|
||||
|
||||
.. _amd-pytorch-training-supported-training-modes-v2510:
|
||||
|
||||
The following table lists supported training modes per model.
|
||||
|
||||
.. datatemplate:yaml:: /data/how-to/rocm-for-ai/training/pytorch-training-benchmark-models.yaml
|
||||
|
||||
{% set model_groups = data.model_groups %}
|
||||
.. dropdown:: Supported training modes
|
||||
|
||||
.. list-table::
|
||||
:header-rows: 1
|
||||
|
||||
* - Model
|
||||
- Supported training modes
|
||||
|
||||
{% for model_group in model_groups %}
|
||||
{% set models = model_group.models %}
|
||||
{% for model in models %}
|
||||
{% if model.training_modes %}
|
||||
* - {{ model.model }}
|
||||
- ``{{ model.training_modes | join('``, ``') }}``
|
||||
|
||||
{% endif %}
|
||||
{% endfor %}
|
||||
{% endfor %}
|
||||
|
||||
.. note::
|
||||
|
||||
Some model and fine-tuning combinations are not listed. This is
|
||||
because the `upstream torchtune repository <https://github.com/pytorch/torchtune>`__
|
||||
doesn't provide default YAML configurations for them.
|
||||
For advanced usage, you can create a custom configuration to enable
|
||||
unlisted fine-tuning methods by using an existing file in the
|
||||
``/workspace/torchtune/recipes/configs`` directory as a template.
|
||||
|
||||
.. _amd-pytorch-training-performance-measurements-v2510:
|
||||
|
||||
Performance measurements
|
||||
========================
|
||||
|
||||
To evaluate performance, the
|
||||
`Performance results with AMD ROCm software <https://www.amd.com/en/developer/resources/rocm-hub/dev-ai/performance-results.html#tabs-a8deaeb413-item-21cea50186-tab>`_
|
||||
page provides reference throughput and latency measurements for training
|
||||
popular AI models.
|
||||
|
||||
.. note::
|
||||
|
||||
The performance data presented in
|
||||
`Performance results with AMD ROCm software <https://www.amd.com/en/developer/resources/rocm-hub/dev-ai/performance-results.html#tabs-a8deaeb413-item-21cea50186-tab>`_
|
||||
should not be interpreted as the peak performance achievable by AMD
|
||||
Instinct MI325X and MI300X GPUs or ROCm software.
|
||||
|
||||
System validation
|
||||
=================
|
||||
|
||||
Before running AI workloads, it's important to validate that your AMD hardware is configured
|
||||
correctly and performing optimally.
|
||||
|
||||
If you have already validated your system settings, including aspects like NUMA auto-balancing, you
|
||||
can skip this step. Otherwise, complete the procedures in the :ref:`System validation and
|
||||
optimization <rocm-for-ai-system-optimization>` guide to properly configure your system settings
|
||||
before starting training.
|
||||
|
||||
To test for optimal performance, consult the recommended :ref:`System health benchmarks
|
||||
<rocm-for-ai-system-health-bench>`. This suite of tests will help you verify and fine-tune your
|
||||
system's configuration.
|
||||
|
||||
This Docker image is optimized for specific model configurations outlined
|
||||
below. Performance can vary for other training workloads, as AMD
|
||||
doesn’t test configurations and run conditions outside those described.
|
||||
|
||||
Run training
|
||||
============
|
||||
|
||||
.. datatemplate:yaml:: /data/how-to/rocm-for-ai/training/pytorch-training-benchmark-models.yaml
|
||||
|
||||
{% set docker = data.docker %}
|
||||
{% set model_groups = data.model_groups %}
|
||||
|
||||
Once the setup is complete, choose between two options to start benchmarking training:
|
||||
|
||||
.. tab-set::
|
||||
|
||||
.. tab-item:: MAD-integrated benchmarking
|
||||
|
||||
{% for model_group in model_groups %}
|
||||
{% for model in model_group.models %}
|
||||
|
||||
.. container:: model-doc {{ model.mad_tag }}
|
||||
|
||||
The following run command is tailored to {{ model.model }}.
|
||||
See :ref:`amd-pytorch-training-model-support-v2510` to switch to another available model.
|
||||
|
||||
1. Clone the ROCm Model Automation and Dashboarding (`<https://github.com/ROCm/MAD>`__) repository to a local
|
||||
directory and install the required packages on the host machine.
|
||||
|
||||
.. code-block:: shell
|
||||
|
||||
git clone https://github.com/ROCm/MAD
|
||||
cd MAD
|
||||
pip install -r requirements.txt
|
||||
|
||||
2. For example, use this command to run the performance benchmark test on the {{ model.model }} model
|
||||
using one node with the {{ model.precision }} data type on the host machine.
|
||||
|
||||
.. code-block:: shell
|
||||
|
||||
export MAD_SECRETS_HFTOKEN="your personal Hugging Face token to access gated models"
|
||||
madengine run \
|
||||
--tags {{ model.mad_tag }} \
|
||||
--keep-model-dir \
|
||||
--live-output \
|
||||
--timeout 28800
|
||||
|
||||
MAD launches a Docker container with the name
|
||||
``container_ci-{{ model.mad_tag }}``. The latency and throughput reports of the
|
||||
model are collected in ``~/MAD/perf.csv``.
|
||||
|
||||
{% endfor %}
|
||||
{% endfor %}
|
||||
|
||||
.. tab-item:: Standalone benchmarking
|
||||
|
||||
{% for model_group in model_groups %}
|
||||
{% for model in model_group.models %}
|
||||
|
||||
.. container:: model-doc {{ model.mad_tag }}
|
||||
|
||||
The following commands are tailored to {{ model.model }}.
|
||||
See :ref:`amd-pytorch-training-model-support-v2510` to switch to another available model.
|
||||
|
||||
{% endfor %}
|
||||
{% endfor %}
|
||||
|
||||
.. rubric:: Download the Docker image and required packages
|
||||
|
||||
1. Use the following command to pull the Docker image from Docker Hub.
|
||||
|
||||
.. code-block:: shell
|
||||
|
||||
docker pull {{ docker.pull_tag }}
|
||||
|
||||
2. Launch the Docker container.
|
||||
|
||||
.. code-block:: shell
|
||||
|
||||
docker run -it \
|
||||
--device /dev/dri \
|
||||
--device /dev/kfd \
|
||||
--network host \
|
||||
--ipc host \
|
||||
--group-add video \
|
||||
--cap-add SYS_PTRACE \
|
||||
--security-opt seccomp=unconfined \
|
||||
--privileged \
|
||||
-v $HOME:$HOME \
|
||||
-v $HOME/.ssh:/root/.ssh \
|
||||
--shm-size 64G \
|
||||
--name training_env \
|
||||
{{ docker.pull_tag }}
|
||||
|
||||
Use these commands if you exit the ``training_env`` container and need to return to it.
|
||||
|
||||
.. code-block:: shell
|
||||
|
||||
docker start training_env
|
||||
docker exec -it training_env bash
|
||||
|
||||
3. In the Docker container, clone the `<https://github.com/ROCm/MAD>`__
|
||||
repository and navigate to the benchmark scripts directory
|
||||
``/workspace/MAD/scripts/pytorch_train``.
|
||||
|
||||
.. code-block:: shell
|
||||
|
||||
git clone https://github.com/ROCm/MAD
|
||||
cd MAD/scripts/pytorch_train
|
||||
|
||||
.. rubric:: Prepare training datasets and dependencies
|
||||
|
||||
1. The following benchmarking examples require downloading models and datasets
|
||||
from Hugging Face. To ensure successful access to gated repos, set your
|
||||
``HF_TOKEN``.
|
||||
|
||||
.. code-block:: shell
|
||||
|
||||
export HF_TOKEN=$your_personal_hugging_face_access_token
|
||||
|
||||
2. Run the setup script to install libraries and datasets needed for benchmarking.
|
||||
|
||||
.. code-block:: shell
|
||||
|
||||
./pytorch_benchmark_setup.sh
|
||||
|
||||
.. container:: model-doc pyt_train_llama-3.1-8b
|
||||
|
||||
``pytorch_benchmark_setup.sh`` installs the following libraries for Llama 3.1 8B:
|
||||
|
||||
.. list-table::
|
||||
:header-rows: 1
|
||||
|
||||
* - Library
|
||||
- Reference
|
||||
|
||||
* - ``accelerate``
|
||||
- `Hugging Face Accelerate <https://huggingface.co/docs/accelerate/en/index>`_
|
||||
|
||||
* - ``datasets``
|
||||
- `Hugging Face Datasets <https://huggingface.co/docs/datasets/v3.2.0/en/index>`_ 3.2.0
|
||||
|
||||
.. container:: model-doc pyt_train_llama-3.1-70b
|
||||
|
||||
``pytorch_benchmark_setup.sh`` installs the following libraries for Llama 3.1 70B:
|
||||
|
||||
.. list-table::
|
||||
:header-rows: 1
|
||||
|
||||
* - Library
|
||||
- Reference
|
||||
|
||||
* - ``datasets``
|
||||
- `Hugging Face Datasets <https://huggingface.co/docs/datasets/v3.2.0/en/index>`_ 3.2.0
|
||||
|
||||
* - ``torchdata``
|
||||
- `TorchData <https://meta-pytorch.org/data/beta/index.html#torchdata>`__
|
||||
|
||||
* - ``tomli``
|
||||
- `Tomli <https://pypi.org/project/tomli/>`__
|
||||
|
||||
* - ``tiktoken``
|
||||
- `tiktoken <https://github.com/openai/tiktoken>`__
|
||||
|
||||
* - ``blobfile``
|
||||
- `blobfile <https://pypi.org/project/blobfile/>`__
|
||||
|
||||
* - ``tabulate``
|
||||
- `tabulate <https://pypi.org/project/tabulate/>`__
|
||||
|
||||
* - ``wandb``
|
||||
- `Weights & Biases <https://github.com/wandb/wandb>`__
|
||||
|
||||
* - ``sentencepiece``
|
||||
- `SentencePiece <https://github.com/google/sentencepiece>`__ 0.2.0
|
||||
|
||||
* - ``tensorboard``
|
||||
- `TensorBoard <https://www.tensorflow.org/tensorboard>`__ 2.18.0
|
||||
|
||||
.. container:: model-doc pyt_train_flux
|
||||
|
||||
``pytorch_benchmark_setup.sh`` installs the following libraries for FLUX:
|
||||
|
||||
.. list-table::
|
||||
:header-rows: 1
|
||||
|
||||
* - Library
|
||||
- Reference
|
||||
|
||||
* - ``accelerate``
|
||||
- `Hugging Face Accelerate <https://huggingface.co/docs/accelerate/en/index>`_
|
||||
|
||||
* - ``datasets``
|
||||
- `Hugging Face Datasets <https://huggingface.co/docs/datasets/v3.2.0/en/index>`__ 3.2.0
|
||||
|
||||
* - ``sentencepiece``
|
||||
- `SentencePiece <https://github.com/google/sentencepiece>`__ 0.2.0
|
||||
|
||||
* - ``tensorboard``
|
||||
- `TensorBoard <https://www.tensorflow.org/tensorboard>`__ 2.18.0
|
||||
|
||||
* - ``csvkit``
|
||||
- `csvkit <https://csvkit.readthedocs.io/en/latest/>`__ 2.0.1
|
||||
|
||||
* - ``deepspeed``
|
||||
- `DeepSpeed <https://github.com/deepspeedai/DeepSpeed>`__ 0.16.2
|
||||
|
||||
* - ``diffusers``
|
||||
- `Hugging Face Diffusers <https://huggingface.co/docs/diffusers/en/index>`__ 0.31.0
|
||||
|
||||
* - ``GitPython``
|
||||
- `GitPython <https://github.com/gitpython-developers/GitPython>`__ 3.1.44
|
||||
|
||||
* - ``opencv-python-headless``
|
||||
- `opencv-python-headless <https://pypi.org/project/opencv-python-headless/>`__ 4.10.0.84
|
||||
|
||||
* - ``peft``
|
||||
- `PEFT <https://huggingface.co/docs/peft/en/index>`__ 0.14.0
|
||||
|
||||
* - ``protobuf``
|
||||
- `Protocol Buffers <https://github.com/protocolbuffers/protobuf>`__ 5.29.2
|
||||
|
||||
* - ``pytest``
|
||||
- `PyTest <https://docs.pytest.org/en/stable/>`__ 8.3.4
|
||||
|
||||
* - ``python-dotenv``
|
||||
- `python-dotenv <https://pypi.org/project/python-dotenv/>`__ 1.0.1
|
||||
|
||||
* - ``seaborn``
|
||||
- `Seaborn <https://seaborn.pydata.org/>`__ 0.13.2
|
||||
|
||||
* - ``transformers``
|
||||
- `Transformers <https://huggingface.co/docs/transformers/en/index>`__ 4.47.0
|
||||
|
||||
``pytorch_benchmark_setup.sh`` downloads the following datasets from Hugging Face:
|
||||
|
||||
* `frank-chieng/chinese_architecture_siheyuan <https://huggingface.co/datasets/frank-chieng/chinese_architecture_siheyuan>`__
|
||||
|
||||
{% for model_group in model_groups %}
|
||||
{% for model in model_group.models %}
|
||||
{% set training_modes = model.training_modes %}
|
||||
{% set training_mode_descs = {
|
||||
"pretrain": "Benchmark pre-training.",
|
||||
"HF_pretrain": "Llama 3.1 8B pre-training with FP8 precision."
|
||||
} %}
|
||||
{% set available_modes = training_modes | select("in", ["pretrain", "HF_pretrain"]) | list %}
|
||||
{% if available_modes %}
|
||||
|
||||
.. container:: model-doc {{ model.mad_tag }}
|
||||
|
||||
.. rubric:: Pretraining
|
||||
|
||||
To start the pre-training benchmark, use the following command with the
|
||||
appropriate options. See the following list of options and their descriptions.
|
||||
|
||||
{% if model.mad_tag == "pyt_train_dlrm" %}
|
||||
|
||||
1. Go to the DLRM directory.
|
||||
|
||||
.. code-block:: shell
|
||||
|
||||
cd /workspace/DLRMBenchmark
|
||||
|
||||
2. To run the single node training benchmark for DLRM-v2 with TF32 precision,
|
||||
run the following script.
|
||||
|
||||
.. code-block:: shell
|
||||
|
||||
./launch_training_single_node.sh
|
||||
|
||||
To run with MAD within the Docker container, use the following command.
|
||||
|
||||
.. code-block:: shell
|
||||
|
||||
./pytorch_benchmark_report.sh -t pretrain -m DLRM
|
||||
|
||||
{% else %}
|
||||
|
||||
.. code-block:: shell
|
||||
|
||||
./pytorch_benchmark_report.sh -t {% if available_modes | length == 1 %}{{ available_modes[0] }}{% else %}$training_mode{% endif %} \
|
||||
-m {{ model.model_repo }} \
|
||||
-p $datatype \
|
||||
-s $sequence_length
|
||||
|
||||
{% if model.mad_tag == "pyt_train_flux" %}
|
||||
.. container:: model-doc {{ model.mad_tag }}
|
||||
|
||||
.. note::
|
||||
|
||||
Currently, FLUX models are not supported out-of-the-box on this Docker.
|
||||
To use FLUX, refer to ``rocm/pytorch-training`` Docker: :doc:`pytorch-training-v25.6`
|
||||
|
||||
Occasionally, downloading the Flux dataset might fail. In the event of this
|
||||
error, manually download it from Hugging Face at
|
||||
`black-forest-labs/FLUX.1-dev <https://huggingface.co/black-forest-labs/FLUX.1-dev>`_
|
||||
and save it to `/workspace/FluxBenchmark`. This ensures that the test script can access
|
||||
the required dataset.
|
||||
{% endif %}
|
||||
|
||||
.. list-table::
|
||||
:header-rows: 1
|
||||
|
||||
* - Name
|
||||
- Options
|
||||
- Description
|
||||
|
||||
{% for mode in available_modes %}
|
||||
* - {% if loop.first %}``$training_mode``{% endif %}
|
||||
- ``{{ mode }}``
|
||||
- {{ training_mode_descs[mode] }}
|
||||
{% endfor %}
|
||||
|
||||
* - ``$datatype``
|
||||
- ``BF16``{% if model.mad_tag == "pyt_train_llama-3.1-8b" %} or ``FP8``{% endif %}
|
||||
- Only Llama 3.1 8B supports FP8 precision.
|
||||
|
||||
* - ``$sequence_length``
|
||||
- Sequence length for the language model.
|
||||
- Between 2048 and 8192. 8192 by default.
|
||||
{% endif %}
|
||||
{% endif %}
|
||||
|
||||
{% set training_modes = model.training_modes %}
|
||||
{% set training_mode_descs = {
|
||||
"posttrain": "Benchmark post-training.",
|
||||
} %}
|
||||
{% set available_modes = training_modes | select("in", ["posttrain"]) | list %}
|
||||
{% if available_modes %}
|
||||
|
||||
.. container:: model-doc {{ model.mad_tag }}
|
||||
|
||||
.. rubric:: Post-training
|
||||
|
||||
To start the post-training benchmark, use the following command with the
|
||||
appropriate options. See the following list of options and their descriptions.
|
||||
|
||||
.. code-block:: shell
|
||||
|
||||
./pytorch_benchmark_report.sh -t {% if available_modes | length == 1 %}{{ available_modes[0] }}{% else %}$training_mode{% endif %} \
|
||||
-m {{ model.model_repo }} \
|
||||
-p $datatype \
|
||||
-s $sequence_length
|
||||
|
||||
.. list-table::
|
||||
:header-rows: 1
|
||||
|
||||
* - Name
|
||||
- Options
|
||||
- Description
|
||||
|
||||
{% for mode in available_modes %}
|
||||
* - {% if loop.first %}``$training_mode``{% endif %}
|
||||
- ``{{ mode }}``
|
||||
- {{ training_mode_descs[mode] }}
|
||||
{% endfor %}
|
||||
|
||||
* - ``$datatype``
|
||||
- ``BF16``{% if model.mad_tag == "pyt_train_llama-3.1-8b" %} or ``FP8``{% endif %}
|
||||
- Only Llama 3.1 8B supports FP8 precision.
|
||||
|
||||
* - ``$sequence_length``
|
||||
- Sequence length for the language model.
|
||||
- Between 2048 and 8192. 8192 by default.
|
||||
{% endif %}
|
||||
|
||||
{% set training_mode_descs = {
|
||||
"finetune_fw": "Full weight fine-tuning (BF16 and FP8 supported).",
|
||||
"finetune_lora": "LoRA fine-tuning (BF16 supported).",
|
||||
"finetune_qlora": "QLoRA fine-tuning (BF16 supported).",
|
||||
"HF_finetune_lora": "LoRA fine-tuning with Hugging Face PEFT.",
|
||||
} %}
|
||||
{% set available_modes = training_modes | select("in", ["finetune_fw", "finetune_lora", "finetune_qlora", "HF_finetune_lora"]) | list %}
|
||||
{% if available_modes %}
|
||||
.. container:: model-doc {{ model.mad_tag }}
|
||||
|
||||
.. rubric:: Fine-tuning
|
||||
|
||||
To start the fine-tuning benchmark, use the following command with the
|
||||
appropriate options. See the following list of options and their descriptions.
|
||||
See :ref:`supported training modes <amd-pytorch-training-supported-training-modes-v2510>`.
|
||||
|
||||
.. code-block:: shell
|
||||
|
||||
./pytorch_benchmark_report.sh -t $training_mode \
|
||||
-m {{ model.model_repo }} \
|
||||
-p $datatype \
|
||||
-s $sequence_length
|
||||
|
||||
.. list-table::
|
||||
:header-rows: 1
|
||||
|
||||
* - Name
|
||||
- Options
|
||||
- Description
|
||||
|
||||
{% for mode in available_modes %}
|
||||
* - {% if loop.first %}``$training_mode``{% endif %}
|
||||
- ``{{ mode }}``
|
||||
- {{ training_mode_descs[mode] }}
|
||||
{% endfor %}
|
||||
|
||||
* - ``$datatype``
|
||||
- ``BF16``{% if "finetune_fw" in available_modes %} or ``FP8``{% endif %}
|
||||
- All models support BF16.{% if "finetune_fw" in available_modes %} FP8 is only available for full weight fine-tuning.{% endif %}
|
||||
|
||||
* - ``$sequence_length``
|
||||
- Between 2048 and 16384.
|
||||
- Sequence length for the language model.
|
||||
|
||||
{% if model.mad_tag in ["pyt_train_llama3.2-vision-11b", "pyt_train_llama-3.2-vision-90b"] %}
|
||||
.. note::
|
||||
|
||||
For LoRA and QLoRA support with vision models (Llama 3.2 11B and 90B),
|
||||
use the following torchtune commit for compatibility:
|
||||
|
||||
.. code-block:: shell
|
||||
|
||||
git checkout 48192e23188b1fc524dd6d127725ceb2348e7f0e
|
||||
|
||||
{% elif model.mad_tag in ["pyt_train_llama-2-7b", "pyt_train_llama-2-13b", "pyt_train_llama-2-70b"] %}
|
||||
.. note::
|
||||
|
||||
You might encounter the following error with Llama 2: ``ValueError: seq_len (16384) of
|
||||
input tensor should be smaller than max_seq_len (4096)``.
|
||||
This error indicates that an input sequence is longer than the model's maximum context window.
|
||||
|
||||
Ensure your tokenized input does not exceed the model's ``max_seq_len`` (4096
|
||||
tokens in this case). You can resolve this by truncating the input or splitting
|
||||
it into smaller chunks before passing it to the model.
|
||||
|
||||
Note on reproducibility: The results in this guide are based on
|
||||
commit ``b4c98ac`` from the upstream
|
||||
`<https://github.com/pytorch/torchtune>`__ repository. For the
|
||||
latest updates, you can use the main branch.
|
||||
|
||||
{% endif %}
|
||||
{% endif %}
|
||||
{% endfor %}
|
||||
{% endfor %}
|
||||
|
||||
.. rubric:: Benchmarking examples
|
||||
|
||||
For examples of benchmarking commands, see `<https://github.com/ROCm/MAD/tree/develop/benchmark/pytorch_train#benchmarking-examples>`__.
|
||||
|
||||
.. _amd-pytorch-training-multinode-examples-v2510:
|
||||
|
||||
Multi-node training
|
||||
-------------------
|
||||
|
||||
Refer to :doc:`/how-to/rocm-for-ai/system-setup/multi-node-setup` to configure your environment for multi-node
|
||||
training. See :ref:`rocm-for-ai-multi-node-setup-pyt-train-example` for example Slurm run commands.
|
||||
|
||||
Pre-training
|
||||
~~~~~~~~~~~~
|
||||
|
||||
Multi-node training with torchtitan is supported. The provided SLURM script is pre-configured for Llama 3 70B.
|
||||
|
||||
To launch the training job on a SLURM cluster for Llama 3 70B, run the following commands from the MAD repository.
|
||||
|
||||
.. code-block:: shell
|
||||
|
||||
# In the MAD repository
|
||||
cd scripts/pytorch_train
|
||||
sbatch run_slurm_train.sh
|
||||
|
||||
Fine-tuning
|
||||
~~~~~~~~~~~
|
||||
|
||||
Multi-node training with torchtune is supported. The provided SLURM script is pre-configured for Llama 3.3 70B.
|
||||
|
||||
To launch the training job on a SLURM cluster for Llama 3.3 70B, run the following commands from the MAD repository.
|
||||
|
||||
.. code-block:: shell
|
||||
|
||||
huggingface-cli login # Get access to HF Llama model space
|
||||
huggingface-cli download meta-llama/Llama-3.3-70B-Instruct --local-dir ./models/Llama-3.3-70B-Instruct # Download the Llama 3.3 model locally
|
||||
# In the MAD repository
|
||||
cd scripts/pytorch_train
|
||||
sbatch Torchtune_Multinode.sh
|
||||
|
||||
.. note::
|
||||
|
||||
Information regarding benchmark setup:
|
||||
|
||||
* By default, Llama 3.3 70B is fine-tuned using ``alpaca_dataset``.
|
||||
* You can adjust the torchtune `YAML configuration file
|
||||
<https://github.com/pytorch/torchtune/blob/main/recipes/configs/llama3_3/70B_full_multinode.yaml>`__
|
||||
if you're using a different model.
|
||||
* The number of nodes and other parameters can be tuned in the SLURM script ``Torchtune_Multinode.sh``.
|
||||
* Set the ``mounting_paths`` inside the SLURM script.
|
||||
|
||||
Once the run is finished, you can find the log files in the ``result_torchtune/`` directory.
|
||||
|
||||
Further reading
|
||||
===============
|
||||
|
||||
- To learn more about MAD and the ``madengine`` CLI, see the `MAD usage guide <https://github.com/ROCm/MAD?tab=readme-ov-file#usage-guide>`__.
|
||||
|
||||
- To learn more about system settings and management practices to configure your system for
|
||||
AMD Instinct MI300X Series GPUs, see `AMD Instinct MI300X system optimization <https://instinct.docs.amd.com/projects/amdgpu-docs/en/latest/system-optimization/mi300x.html>`_.
|
||||
|
||||
- For a list of other ready-made Docker images for AI with ROCm, see
|
||||
`AMD Infinity Hub <https://www.amd.com/en/developer/resources/infinity-hub.html#f-amd_hub_category=AI%20%26%20ML%20Models>`_.
|
||||
|
||||
Previous versions
|
||||
=================
|
||||
|
||||
See :doc:`pytorch-training-history` to find documentation for previous releases
|
||||
of the ``ROCm/pytorch-training`` Docker image.
|
||||
@@ -47,7 +47,7 @@ Megatron-LM.
|
||||
- {{ component_version }}
|
||||
{% endfor %}
|
||||
|
||||
.. _amd-primus-megatron-lm-model-support-v2510:
|
||||
.. _amd-primus-megatron-lm-model-support-v25.11:
|
||||
|
||||
Supported models
|
||||
================
|
||||
@@ -108,7 +108,7 @@ To test for optimal performance, consult the recommended :ref:`System health ben
|
||||
<rocm-for-ai-system-health-bench>`. This suite of tests will help you verify and fine-tune your
|
||||
system's configuration.
|
||||
|
||||
.. _mi300x-amd-primus-megatron-lm-training-v2510:
|
||||
.. _mi300x-amd-primus-megatron-lm-training-v25.11:
|
||||
|
||||
Environment setup
|
||||
=================
|
||||
@@ -118,7 +118,7 @@ Environment setup
|
||||
Use the following instructions to set up the environment, configure the script to train models, and
|
||||
reproduce the benchmark results on AMD Instinct GPUs.
|
||||
|
||||
.. _amd-primus-megatron-lm-requirements-v2510:
|
||||
.. _amd-primus-megatron-lm-requirements-v25.11:
|
||||
|
||||
Pull the Docker image
|
||||
|
||||
@@ -157,16 +157,16 @@ Pull the Docker image
|
||||
docker start primus_training_env
|
||||
docker exec -it primus_training_env bash
|
||||
|
||||
The Docker container hosts verified branch ``release/v25.10`` of the `Primus
|
||||
<https://github.com/AMD-AGI/Primus/tree/release/v25.10>`__ repository.
|
||||
The Docker container hosts verified commit ``c4c083de`` of the `Primus
|
||||
<https://github.com/AMD-AGI/Primus/tree/c4c083de64ba3e8f19ccc9629411267108931f9e/>`__ repository.
|
||||
|
||||
.. _amd-primus-megatron-lm-environment-setup-v2510:
|
||||
.. _amd-primus-megatron-lm-environment-setup-v25.11:
|
||||
|
||||
Configuration
|
||||
=============
|
||||
|
||||
Primus defines a training configuration in YAML for each model in
|
||||
`examples/megatron/configs <https://github.com/AMD-AGI/Primus/tree/e16b27bf6c1b2798f38848fc574fee60d9a9b902/examples/megatron/configs>`__.
|
||||
`examples/megatron/configs <https://github.com/AMD-AGI/Primus/tree/c4c083de64ba3e8f19ccc9629411267108931f9e/examples/megatron/configs>`__.
|
||||
|
||||
.. datatemplate:yaml:: /data/how-to/rocm-for-ai/training/primus-megatron-benchmark-models.yaml
|
||||
|
||||
@@ -207,7 +207,7 @@ You can use either mock data or real data for training.
|
||||
|
||||
Ensure that the files are accessible inside the Docker container.
|
||||
|
||||
.. _amd-primus-megatron-lm-tokenizer-v2510:
|
||||
.. _amd-primus-megatron-lm-tokenizer-v25.11:
|
||||
|
||||
Tokenizer
|
||||
---------
|
||||
@@ -228,7 +228,7 @@ right permissions to access the tokenizer for each model.
|
||||
<https://github.com/AMD-AGI/Primus/blob/e16b27bf6c1b2798f38848fc574fee60d9a9b902/examples/megatron/configs/llama3.1_8B-pretrain.yaml>`__
|
||||
definition.
|
||||
|
||||
.. _amd-primus-megatron-lm-run-training-v2510:
|
||||
.. _amd-primus-megatron-lm-run-training-v25.11:
|
||||
|
||||
Run training
|
||||
============
|
||||
@@ -252,7 +252,7 @@ To run training on a single node, navigate to ``/workspace/Primus`` and use the
|
||||
|
||||
Once setup is complete, run the appropriate training command.
|
||||
The following run commands are tailored to Llama 3.3 70B.
|
||||
See :ref:`amd-primus-megatron-lm-model-support-v2510` to switch to another available model.
|
||||
See :ref:`amd-primus-megatron-lm-model-support-v25.11` to switch to another available model.
|
||||
|
||||
To run pre-training for Llama 3.3 70B BF16, run:
|
||||
|
||||
@@ -263,11 +263,8 @@ To run training on a single node, navigate to ``/workspace/Primus`` and use the
|
||||
|
||||
.. code-block:: shell
|
||||
|
||||
EXP=examples/megatron/configs/MI355X/llama3.3_70B-pretrain.yaml \
|
||||
bash ./examples/run_pretrain.sh \
|
||||
--train_iters 50 \
|
||||
--micro_batch_size 6 \
|
||||
--global_batch_size 48 \
|
||||
EXP=examples/megatron/configs/MI355X/llama3.3_70B-BF16-pretrain.yaml \
|
||||
bash ./examples/run_pretrain.sh
|
||||
|
||||
.. tab-item:: MI300X
|
||||
:sync: MI325X and MI300X
|
||||
@@ -279,17 +276,14 @@ To run training on a single node, navigate to ``/workspace/Primus`` and use the
|
||||
export PRIMUS_TURBO_ATTN_V3_ATOMIC_FP32=1
|
||||
export NVTE_CK_IS_V3_ATOMIC_FP32=1
|
||||
|
||||
EXP=examples/megatron/configs/MI300X/llama3.3_70B-pretrain.yaml \
|
||||
bash ./examples/run_pretrain.sh \
|
||||
--train_iters 50 \
|
||||
--micro_batch_size 2 \
|
||||
--global_batch_size 16
|
||||
EXP=examples/megatron/configs/MI300X/llama3.3_70B-BF16-pretrain.yaml \
|
||||
bash ./examples/run_pretrain.sh
|
||||
|
||||
.. container:: model-doc primus_pyt_megatron_lm_train_llama-3.1-8b
|
||||
|
||||
Once setup is complete, run the appropriate training command.
|
||||
The following run commands are tailored to Llama 3.1 8B.
|
||||
See :ref:`amd-primus-megatron-lm-model-support-v2510` to switch to another available model.
|
||||
See :ref:`amd-primus-megatron-lm-model-support-v25.11` to switch to another available model.
|
||||
|
||||
To run pre-training for Llama 3.1 8B FP8, run:
|
||||
|
||||
@@ -300,12 +294,8 @@ To run training on a single node, navigate to ``/workspace/Primus`` and use the
|
||||
|
||||
.. code-block:: shell
|
||||
|
||||
EXP=examples/megatron/configs/MI355X/llama3.1_8B-pretrain.yaml \
|
||||
bash ./examples/run_pretrain.sh \
|
||||
--train_iters 50 \
|
||||
--fp8 hybrid \
|
||||
--micro_batch_size 4 \
|
||||
--global_batch_size 512 \
|
||||
EXP=examples/megatron/configs/MI355X/llama3.1_8B-FP8-pretrain.yaml \
|
||||
bash ./examples/run_pretrain.sh
|
||||
|
||||
.. tab-item:: MI300X
|
||||
:sync: MI325X and MI300X
|
||||
@@ -317,10 +307,8 @@ To run training on a single node, navigate to ``/workspace/Primus`` and use the
|
||||
export PRIMUS_TURBO_ATTN_V3_ATOMIC_FP32=1
|
||||
export NVTE_CK_IS_V3_ATOMIC_FP32=1
|
||||
|
||||
EXP=examples/megatron/configs/MI300X/llama3.1_8B-pretrain.yaml \
|
||||
bash ./examples/run_pretrain.sh \
|
||||
--train_iters 50 \
|
||||
--fp8 hybrid
|
||||
EXP=examples/megatron/configs/MI300X/llama3.1_8B-FP8-pretrain.yaml \
|
||||
bash ./examples/run_pretrain.sh
|
||||
|
||||
For Llama 3.1 8B BF16, use the following command:
|
||||
|
||||
@@ -331,11 +319,8 @@ To run training on a single node, navigate to ``/workspace/Primus`` and use the
|
||||
|
||||
.. code-block:: shell
|
||||
|
||||
EXP=examples/megatron/configs/MI355X/llama3.1_8B-pretrain.yaml \
|
||||
bash ./examples/run_pretrain.sh \
|
||||
--train_iters 50 \
|
||||
--micro_batch_size 4 \
|
||||
--global_batch_size 512 \
|
||||
EXP=examples/megatron/configs/MI355X/llama3.1_BF16-pretrain.yaml \
|
||||
bash ./examples/run_pretrain.sh
|
||||
|
||||
.. tab-item:: MI300X
|
||||
:sync: MI325X and MI300X
|
||||
@@ -347,15 +332,14 @@ To run training on a single node, navigate to ``/workspace/Primus`` and use the
|
||||
export PRIMUS_TURBO_ATTN_V3_ATOMIC_FP32=1
|
||||
export NVTE_CK_IS_V3_ATOMIC_FP32=1
|
||||
|
||||
EXP=examples/megatron/configs/MI300X/llama3.1_8B-pretrain.yaml \
|
||||
bash ./examples/run_pretrain.sh \
|
||||
--train_iters 50
|
||||
EXP=examples/megatron/configs/MI300X/llama3.1_8B-BF16-pretrain.yaml \
|
||||
bash ./examples/run_pretrain.sh
|
||||
|
||||
.. container:: model-doc primus_pyt_megatron_lm_train_llama-3.1-70b
|
||||
|
||||
Once setup is complete, run the appropriate training command.
|
||||
The following run commands are tailored to Llama 3.1 70B.
|
||||
See :ref:`amd-primus-megatron-lm-model-support-v2510` to switch to another available model.
|
||||
See :ref:`amd-primus-megatron-lm-model-support-v25.11` to switch to another available model.
|
||||
|
||||
To run pre-training for Llama 3.1 70B BF16, run:
|
||||
|
||||
@@ -366,11 +350,8 @@ To run training on a single node, navigate to ``/workspace/Primus`` and use the
|
||||
|
||||
.. code-block:: shell
|
||||
|
||||
EXP=examples/megatron/configs/MI355X/llama3.1_70B-pretrain.yaml \
|
||||
bash ./examples/run_pretrain.sh \
|
||||
--train_iters 50 \
|
||||
--micro_batch_size 4 \
|
||||
--global_batch_size 32
|
||||
EXP=examples/megatron/configs/MI355X/llama3.1_70B-BF16-pretrain.yaml \
|
||||
bash ./examples/run_pretrain.sh
|
||||
|
||||
.. tab-item:: MI300X
|
||||
:sync: MI325X and MI300X
|
||||
@@ -382,9 +363,8 @@ To run training on a single node, navigate to ``/workspace/Primus`` and use the
|
||||
export PRIMUS_TURBO_ATTN_V3_ATOMIC_FP32=1
|
||||
export NVTE_CK_IS_V3_ATOMIC_FP32=1
|
||||
|
||||
EXP=examples/megatron/configs/MI300X/llama3.1_70B-pretrain.yaml \
|
||||
bash ./examples/run_pretrain.sh \
|
||||
--train_iters 50
|
||||
EXP=examples/megatron/configs/MI300X/llama3.1_70B-BF16-pretrain.yaml \
|
||||
bash ./examples/run_pretrain.sh
|
||||
|
||||
To run the training on a single node for Llama 3.1 70B FP8, use the following command.
|
||||
|
||||
@@ -401,13 +381,8 @@ To run training on a single node, navigate to ``/workspace/Primus`` and use the
|
||||
|
||||
.. code-block:: shell
|
||||
|
||||
EXP=examples/megatron/configs/MI355X/llama3.1_70B-pretrain.yaml \
|
||||
bash ./examples/run_pretrain.sh \
|
||||
--train_iters 50 \
|
||||
--fp8 hybrid \
|
||||
--no_fp8_weight_transpose_cache true \
|
||||
--micro_batch_size 3 \
|
||||
--global_batch_size 24
|
||||
EXP=examples/megatron/configs/MI355X/llama3.1_70B-FP8-pretrain.yaml \
|
||||
bash ./examples/run_pretrain.sh
|
||||
|
||||
.. tab-item:: MI300X
|
||||
:sync: MI325X and MI300X
|
||||
@@ -419,7 +394,7 @@ To run training on a single node, navigate to ``/workspace/Primus`` and use the
|
||||
export PRIMUS_TURBO_ATTN_V3_ATOMIC_FP32=1
|
||||
export NVTE_CK_IS_V3_ATOMIC_FP32=1
|
||||
|
||||
EXP=examples/megatron/configs/MI300X/llama3.1_70B-pretrain.yaml \
|
||||
EXP=examples/megatron/configs/MI300X/llama3.1_70B-FP8-pretrain.yaml \
|
||||
bash ./examples/run_pretrain.sh \
|
||||
--train_iters 50 \
|
||||
--num_layers 40 \
|
||||
@@ -430,7 +405,7 @@ To run training on a single node, navigate to ``/workspace/Primus`` and use the
|
||||
|
||||
Once setup is complete, run the appropriate training command.
|
||||
The following run commands are tailored to Llama 2 7B.
|
||||
See :ref:`amd-primus-megatron-lm-model-support-v2510` to switch to another available model.
|
||||
See :ref:`amd-primus-megatron-lm-model-support-v25.11` to switch to another available model.
|
||||
|
||||
To run pre-training for Llama 2 7B FP8, run:
|
||||
|
||||
@@ -441,12 +416,8 @@ To run training on a single node, navigate to ``/workspace/Primus`` and use the
|
||||
|
||||
.. code-block:: shell
|
||||
|
||||
EXP=examples/megatron/configs/MI355X/llama2_7B-pretrain.yaml \
|
||||
bash ./examples/run_pretrain.sh \
|
||||
--train_iters 50 \
|
||||
--fp8 hybrid \
|
||||
--micro_batch_size 13 \
|
||||
--global_batch_size 416
|
||||
EXP=examples/megatron/configs/MI355X/llama2_7B-FP8-pretrain.yaml \
|
||||
bash ./examples/run_pretrain.sh
|
||||
|
||||
.. tab-item:: MI300X
|
||||
:sync: MI325X and MI300X
|
||||
@@ -458,10 +429,8 @@ To run training on a single node, navigate to ``/workspace/Primus`` and use the
|
||||
export PRIMUS_TURBO_ATTN_V3_ATOMIC_FP32=1
|
||||
export NVTE_CK_IS_V3_ATOMIC_FP32=1
|
||||
|
||||
EXP=examples/megatron/configs/MI300X/llama2_7B-pretrain.yaml \
|
||||
bash ./examples/run_pretrain.sh \
|
||||
--train_iters 50 \
|
||||
--fp8 hybrid
|
||||
EXP=examples/megatron/configs/MI300X/llama2_7B-FP8-pretrain.yaml \
|
||||
bash ./examples/run_pretrain.sh
|
||||
|
||||
To run pre-training for Llama 2 7B BF16, run:
|
||||
|
||||
@@ -472,11 +441,8 @@ To run training on a single node, navigate to ``/workspace/Primus`` and use the
|
||||
|
||||
.. code-block:: shell
|
||||
|
||||
EXP=examples/megatron/configs/MI355X/llama2_7B-pretrain.yaml \
|
||||
bash ./examples/run_pretrain.sh \
|
||||
--train_iters 50 \
|
||||
--micro_batch_size 10 \
|
||||
--global_batch_size 640
|
||||
EXP=examples/megatron/configs/MI355X/llama2_7B-BF16-pretrain.yaml \
|
||||
bash ./examples/run_pretrain.sh
|
||||
|
||||
.. tab-item:: MI300X
|
||||
:sync: MI325X and MI300X
|
||||
@@ -488,15 +454,14 @@ To run training on a single node, navigate to ``/workspace/Primus`` and use the
|
||||
export PRIMUS_TURBO_ATTN_V3_ATOMIC_FP32=1
|
||||
export NVTE_CK_IS_V3_ATOMIC_FP32=1
|
||||
|
||||
EXP=examples/megatron/configs/MI300X/llama2_7B-pretrain.yaml \
|
||||
bash ./examples/run_pretrain.sh \
|
||||
--train_iters 50
|
||||
EXP=examples/megatron/configs/MI300X/llama2_7B-BF16-pretrain.yaml \
|
||||
bash ./examples/run_pretrain.sh
|
||||
|
||||
.. container:: model-doc primus_pyt_megatron_lm_train_llama-2-70b
|
||||
|
||||
Once setup is complete, run the appropriate training command.
|
||||
The following run commands are tailored to Llama 2 70B.
|
||||
See :ref:`amd-primus-megatron-lm-model-support-v2510` to switch to another available model.
|
||||
See :ref:`amd-primus-megatron-lm-model-support-v25.11` to switch to another available model.
|
||||
|
||||
To run pre-training for Llama 2 70B BF16, run:
|
||||
|
||||
@@ -507,11 +472,8 @@ To run training on a single node, navigate to ``/workspace/Primus`` and use the
|
||||
|
||||
.. code-block:: shell
|
||||
|
||||
EXP=examples/megatron/configs/MI355X/llama2_70B-pretrain.yaml \
|
||||
bash ./examples/run_pretrain.sh \
|
||||
--train_iters 50 \
|
||||
--micro_batch_size 17 \
|
||||
--global_batch_size 272
|
||||
EXP=examples/megatron/configs/MI355X/llama2_70B-BF16-pretrain.yaml \
|
||||
bash ./examples/run_pretrain.sh
|
||||
|
||||
.. tab-item:: MI300X
|
||||
:sync: MI325X and MI300X
|
||||
@@ -523,15 +485,14 @@ To run training on a single node, navigate to ``/workspace/Primus`` and use the
|
||||
export PRIMUS_TURBO_ATTN_V3_ATOMIC_FP32=1
|
||||
export NVTE_CK_IS_V3_ATOMIC_FP32=1
|
||||
|
||||
EXP=examples/megatron/configs/MI300X/llama2_70B-pretrain.yaml \
|
||||
bash ./examples/run_pretrain.sh \
|
||||
--train_iters 50
|
||||
EXP=examples/megatron/configs/MI300X/llama2_70B-BF16-pretrain.yaml \
|
||||
bash ./examples/run_pretrain.sh
|
||||
|
||||
.. container:: model-doc primus_pyt_megatron_lm_train_deepseek-v3-proxy
|
||||
|
||||
Once setup is complete, run the appropriate training command.
|
||||
The following run commands are tailored to DeepSeek-V3.
|
||||
See :ref:`amd-primus-megatron-lm-model-support-v2510` to switch to another available model.
|
||||
See :ref:`amd-primus-megatron-lm-model-support-v25.11` to switch to another available model.
|
||||
|
||||
To run training on a single node for DeepSeek-V3 (MoE with expert parallel) BF16 with 3-layer proxy,
|
||||
use the following command:
|
||||
@@ -543,7 +504,7 @@ To run training on a single node, navigate to ``/workspace/Primus`` and use the
|
||||
|
||||
.. code-block:: shell
|
||||
|
||||
EXP=examples/megatron/configs/MI355X/deepseek_v3-pretrain.yaml \
|
||||
EXP=examples/megatron/configs/MI355X/deepseek_v3-BF16-pretrain.yaml \
|
||||
bash examples/run_pretrain.sh \
|
||||
--num_layers 3 \
|
||||
--moe_layer_freq 1 \
|
||||
@@ -561,19 +522,17 @@ To run training on a single node, navigate to ``/workspace/Primus`` and use the
|
||||
export PRIMUS_TURBO_ATTN_V3_ATOMIC_FP32=1
|
||||
export NVTE_CK_IS_V3_ATOMIC_FP32=1
|
||||
|
||||
EXP=examples/megatron/configs/MI300X/deepseek_v3-pretrain.yaml \
|
||||
EXP=examples/megatron/configs/MI300X/deepseek_v3-BF16-pretrain.yaml \
|
||||
bash examples/run_pretrain.sh \
|
||||
--num_layers 3 \
|
||||
--moe_layer_freq 1 \
|
||||
--micro_batch_size 3 \
|
||||
--global_batch_size 192 \
|
||||
--train_iters 50
|
||||
|
||||
.. container:: model-doc primus_pyt_megatron_lm_train_deepseek-v2-lite-16b
|
||||
|
||||
Once setup is complete, run the appropriate training command.
|
||||
The following run commands are tailored to DeepSeek-V2-Lite.
|
||||
See :ref:`amd-primus-megatron-lm-model-support-v2510` to switch to another available model.
|
||||
See :ref:`amd-primus-megatron-lm-model-support-v25.11` to switch to another available model.
|
||||
|
||||
To run training on a single node for DeepSeek-V2-Lite (MoE with expert parallel) BF16,
|
||||
use the following command:
|
||||
@@ -585,11 +544,8 @@ To run training on a single node, navigate to ``/workspace/Primus`` and use the
|
||||
|
||||
.. code-block:: shell
|
||||
|
||||
EXP=examples/megatron/configs/MI355X/deepseek_v2_lite-pretrain.yaml \
|
||||
bash examples/run_pretrain.sh \
|
||||
--train_iters 50 \
|
||||
--micro_batch_size 12 \
|
||||
--global_batch_size 768
|
||||
EXP=examples/megatron/configs/MI355X/deepseek_v2_lite-BF16-pretrain.yaml \
|
||||
bash examples/run_pretrain.sh
|
||||
|
||||
.. tab-item:: MI300X
|
||||
:sync: MI325X and MI300X
|
||||
@@ -601,16 +557,14 @@ To run training on a single node, navigate to ``/workspace/Primus`` and use the
|
||||
export PRIMUS_TURBO_ATTN_V3_ATOMIC_FP32=1
|
||||
export NVTE_CK_IS_V3_ATOMIC_FP32=1
|
||||
|
||||
EXP=examples/megatron/configs/MI300X/deepseek_v2_lite-pretrain.yaml \
|
||||
bash examples/run_pretrain.sh \
|
||||
--train_iters 50 \
|
||||
--global_batch_size 256
|
||||
EXP=examples/megatron/configs/MI300X/deepseek_v2_lite-BF16-pretrain.yaml \
|
||||
bash examples/run_pretrain.sh
|
||||
|
||||
.. container:: model-doc primus_pyt_megatron_lm_train_mixtral-8x7b
|
||||
|
||||
Once setup is complete, run the appropriate training command.
|
||||
The following run commands are tailored to Mixtral 8x7B.
|
||||
See :ref:`amd-primus-megatron-lm-model-support-v2510` to switch to another available model.
|
||||
See :ref:`amd-primus-megatron-lm-model-support-v25.11` to switch to another available model.
|
||||
|
||||
To run training on a single node for Mixtral 8x7B (MoE with expert parallel),
|
||||
use the following command:
|
||||
@@ -622,11 +576,8 @@ To run training on a single node, navigate to ``/workspace/Primus`` and use the
|
||||
|
||||
.. code-block:: shell
|
||||
|
||||
EXP=examples/megatron/configs/MI355X/mixtral_8x7B_v0.1-pretrain.yaml \
|
||||
bash examples/run_pretrain.sh \
|
||||
--train_iters 50 \
|
||||
--micro_batch_size 4 \
|
||||
--global_batch_size 256
|
||||
EXP=examples/megatron/configs/MI355X/mixtral_8x7B_v0.1-BF16-pretrain.yaml \
|
||||
bash examples/run_pretrain.sh
|
||||
|
||||
.. tab-item:: MI300X
|
||||
:sync: MI325X and MI300X
|
||||
@@ -638,7 +589,7 @@ To run training on a single node, navigate to ``/workspace/Primus`` and use the
|
||||
export PRIMUS_TURBO_ATTN_V3_ATOMIC_FP32=1
|
||||
export NVTE_CK_IS_V3_ATOMIC_FP32=1
|
||||
|
||||
EXP=examples/megatron/configs/MI300X/mixtral_8x7B_v0.1-pretrain.yaml \
|
||||
EXP=examples/megatron/configs/MI300X/mixtral_8x7B_v0.1-BF16-pretrain.yaml \
|
||||
bash examples/run_pretrain.sh \
|
||||
--train_iters 50
|
||||
|
||||
@@ -646,7 +597,7 @@ To run training on a single node, navigate to ``/workspace/Primus`` and use the
|
||||
|
||||
Once setup is complete, run the appropriate training command.
|
||||
The following run commands are tailored to Mixtral 8x22B.
|
||||
See :ref:`amd-primus-megatron-lm-model-support-v2510` to switch to another available model.
|
||||
See :ref:`amd-primus-megatron-lm-model-support-v25.11` to switch to another available model.
|
||||
|
||||
To run training on a single node for Mixtral 8x22B BF16 (MoE with expert parallel) 4-layer proxy,
|
||||
use the following command:
|
||||
@@ -658,13 +609,8 @@ To run training on a single node, navigate to ``/workspace/Primus`` and use the
|
||||
|
||||
.. code-block:: shell
|
||||
|
||||
EXP=examples/megatron/configs/MI355X/mixtral_8x22B_v0.1-pretrain.yaml \
|
||||
bash examples/run_pretrain.sh \
|
||||
--train_iters 50 \
|
||||
--num_layers 4 \
|
||||
--pipeline_model_parallel_size 1 \
|
||||
--micro_batch_size 2 \
|
||||
--global_batch_size 16
|
||||
EXP=examples/megatron/configs/MI355X/mixtral_8x22B_v0.1-BF16-pretrain.yaml \
|
||||
bash examples/run_pretrain.sh
|
||||
|
||||
.. tab-item:: MI300X
|
||||
:sync: MI325X and MI300X
|
||||
@@ -676,7 +622,7 @@ To run training on a single node, navigate to ``/workspace/Primus`` and use the
|
||||
export PRIMUS_TURBO_ATTN_V3_ATOMIC_FP32=1
|
||||
export NVTE_CK_IS_V3_ATOMIC_FP32=1
|
||||
|
||||
EXP=examples/megatron/configs/MI300X/mixtral_8x22B_v0.1-pretrain.yaml \
|
||||
EXP=examples/megatron/configs/MI300X/mixtral_8x22B_v0.1-BF16-pretrain.yaml \
|
||||
bash examples/run_pretrain.sh \
|
||||
--train_iters 50 \
|
||||
--num_layers 4 \
|
||||
@@ -688,7 +634,7 @@ To run training on a single node, navigate to ``/workspace/Primus`` and use the
|
||||
|
||||
Once setup is complete, run the appropriate training command.
|
||||
The following run commands are tailored to Qwen 2.5 7B.
|
||||
See :ref:`amd-primus-megatron-lm-model-support-v2510` to switch to another available model.
|
||||
See :ref:`amd-primus-megatron-lm-model-support-v25.11` to switch to another available model.
|
||||
|
||||
To run training on a single node for Qwen 2.5 7B BF16, use the following
|
||||
command:
|
||||
@@ -700,11 +646,8 @@ To run training on a single node, navigate to ``/workspace/Primus`` and use the
|
||||
|
||||
.. code-block:: shell
|
||||
|
||||
EXP=examples/megatron/configs/MI355X/qwen2.5_7B-pretrain.yaml \
|
||||
bash examples/run_pretrain.sh \
|
||||
--train_iters 50 \
|
||||
--micro_batch_size 16 \
|
||||
--global_batch_size 768
|
||||
EXP=examples/megatron/configs/MI355X/qwen2.5_7B-BF16-pretrain.yaml \
|
||||
bash examples/run_pretrain.sh
|
||||
|
||||
.. tab-item:: MI300X
|
||||
:sync: MI325X and MI300X
|
||||
@@ -716,9 +659,8 @@ To run training on a single node, navigate to ``/workspace/Primus`` and use the
|
||||
export PRIMUS_TURBO_ATTN_V3_ATOMIC_FP32=1
|
||||
export NVTE_CK_IS_V3_ATOMIC_FP32=1
|
||||
|
||||
EXP=examples/megatron/configs/MI300X/qwen2.5_7B-pretrain.yaml \
|
||||
bash examples/run_pretrain.sh \
|
||||
--train_iters 50
|
||||
EXP=examples/megatron/configs/MI300X/qwen2.5_7B-BF16-pretrain.yaml \
|
||||
bash examples/run_pretrain.sh
|
||||
|
||||
For FP8, use the following command.
|
||||
|
||||
@@ -729,12 +671,8 @@ To run training on a single node, navigate to ``/workspace/Primus`` and use the
|
||||
|
||||
.. code-block:: shell
|
||||
|
||||
EXP=examples/megatron/configs/MI355X/qwen2.5_7B-pretrain.yaml \
|
||||
bash examples/run_pretrain.sh \
|
||||
--train_iters 50 \
|
||||
--fp8 hybrid
|
||||
--micro_batch_size 20 \
|
||||
--global_batch_size 800
|
||||
EXP=examples/megatron/configs/MI355X/qwen2.5_7B-FP8-pretrain.yaml \
|
||||
bash examples/run_pretrain.sh
|
||||
|
||||
.. tab-item:: MI300X
|
||||
:sync: MI325X and MI300X
|
||||
@@ -746,16 +684,14 @@ To run training on a single node, navigate to ``/workspace/Primus`` and use the
|
||||
export PRIMUS_TURBO_ATTN_V3_ATOMIC_FP32=1
|
||||
export NVTE_CK_IS_V3_ATOMIC_FP32=1
|
||||
|
||||
EXP=examples/megatron/configs/MI300X/qwen2.5_7B-pretrain.yaml \
|
||||
bash examples/run_pretrain.sh \
|
||||
--train_iters 50 \
|
||||
--fp8 hybrid
|
||||
EXP=examples/megatron/configs/MI300X/qwen2.5_7B-FP8-pretrain.yaml \
|
||||
bash examples/run_pretrain.sh
|
||||
|
||||
.. container:: model-doc primus_pyt_megatron_lm_train_qwen2.5-72b
|
||||
|
||||
Once setup is complete, run the appropriate training command.
|
||||
The following run commands are tailored to Qwen 2.5 72B.
|
||||
See :ref:`amd-primus-megatron-lm-model-support-v2510` to switch to another available model.
|
||||
See :ref:`amd-primus-megatron-lm-model-support-v25.11` to switch to another available model.
|
||||
|
||||
To run the training on a single node for Qwen 2.5 72B BF16, use the following command.
|
||||
|
||||
@@ -782,11 +718,10 @@ To run training on a single node, navigate to ``/workspace/Primus`` and use the
|
||||
export PRIMUS_TURBO_ATTN_V3_ATOMIC_FP32=1
|
||||
export NVTE_CK_IS_V3_ATOMIC_FP32=1
|
||||
|
||||
EXP=examples/megatron/configs/MI300X/qwen2.5_72B-pretrain.yaml \
|
||||
bash examples/run_pretrain.sh \
|
||||
--train_iters 50
|
||||
EXP=examples/megatron/configs/MI300X/qwen2.5_72B-BF16-pretrain.yaml \
|
||||
bash examples/run_pretrain.sh
|
||||
|
||||
.. _amd-primus-megatron-multi-node-examples-v2510:
|
||||
.. _amd-primus-megatron-multi-node-examples-v25.11:
|
||||
|
||||
Multi-node training examples
|
||||
----------------------------
|
||||
@@ -805,7 +740,7 @@ to launch the multi-node workload. Use the following steps to setup your environ
|
||||
|
||||
git clone --recurse-submodules https://github.com/AMD-AGI/Primus.git
|
||||
cd Primus
|
||||
git checkout release/v25.10
|
||||
git checkout c4c083de64ba3e8f19ccc9629411267108931f9e
|
||||
git submodule update --init --recursive
|
||||
|
||||
export DOCKER_IMAGE={{ docker.pull_tag }}
|
||||
@@ -828,13 +763,13 @@ to launch the multi-node workload. Use the following steps to setup your environ
|
||||
* If ``NCCL_IB_HCA`` and ``NCCL_SOCKET_IFNAME`` are not set, Primus will try to auto-detect. However, since NICs can vary accross different cluster, it is encouraged to explicitly export your NCCL parameters for the cluster.
|
||||
* To find your network interface, you can use ``ip a``.
|
||||
* To find RDMA interfaces, you can use ``ibv_devices`` to get the list of all the RDMA/IB devices.
|
||||
* Remember to set ``DOCKER_IMAGE`` and ``HF_TOKEN`` (see :ref:`amd-primus-megatron-lm-tokenizer-v2510`) as appropriate.
|
||||
* Remember to set ``DOCKER_IMAGE`` and ``HF_TOKEN`` (see :ref:`amd-primus-megatron-lm-tokenizer-v25.11`) as appropriate.
|
||||
|
||||
.. container:: model-doc primus_pyt_megatron_lm_train_llama-3.1-8b
|
||||
|
||||
Once setup is complete, run the appropriate training command.
|
||||
The following run commands are tailored to Llama 3.1 8B.
|
||||
See :ref:`amd-primus-megatron-lm-model-support-v2510` to switch to another available model.
|
||||
See :ref:`amd-primus-megatron-lm-model-support-v25.11` to switch to another available model.
|
||||
|
||||
To train Llama 3.1 8B FP8 on 8 nodes, run:
|
||||
|
||||
@@ -843,16 +778,15 @@ to launch the multi-node workload. Use the following steps to setup your environ
|
||||
# Adjust the training parameters.
|
||||
# For example, `global_batch_size: 8 * #single_node_bs` for 8 nodes in this case.
|
||||
NNODES=8 \
|
||||
EXP=examples/megatron/configs/llama3.1_8B-pretrain.yaml \
|
||||
EXP=examples/megatron/configs/MI300X/llama3.1_8B-FP8-pretrain.yaml \
|
||||
bash ./examples/run_slurm_pretrain.sh \
|
||||
--global_batch_size 1024 \
|
||||
--fp8 hybrid
|
||||
|
||||
.. container:: model-doc primus_pyt_megatron_lm_train_llama-2-7b
|
||||
|
||||
Once setup is complete, run the appropriate training command.
|
||||
The following run commands are tailored to Llama 2 7B.
|
||||
See :ref:`amd-primus-megatron-lm-model-support-v2510` to switch to another available model.
|
||||
See :ref:`amd-primus-megatron-lm-model-support-v25.11` to switch to another available model.
|
||||
|
||||
To train Llama 2 7B FP8 on 8 nodes, run:
|
||||
|
||||
@@ -861,16 +795,15 @@ to launch the multi-node workload. Use the following steps to setup your environ
|
||||
# Adjust the training parameters.
|
||||
# For example, `global_batch_size: 8 * #single_node_bs` for 8 nodes in this case.
|
||||
NNODES=8 \
|
||||
EXP=examples/megatron/configs/llama2_7B-pretrain.yaml \
|
||||
EXP=examples/megatron/configs/MI300X/llama2_7B-FP8-pretrain.yaml \
|
||||
bash ./examples/run_slurm_pretrain.sh \
|
||||
--global_batch_size 2048 \
|
||||
--fp8 hybrid
|
||||
|
||||
.. container:: model-doc primus_pyt_megatron_lm_train_llama-3.1-70b
|
||||
|
||||
Once setup is complete, run the appropriate training command.
|
||||
The following run commands are tailored to Llama 3.1 70B.
|
||||
See :ref:`amd-primus-megatron-lm-model-support-v2510` to switch to another available model.
|
||||
See :ref:`amd-primus-megatron-lm-model-support-v25.11` to switch to another available model.
|
||||
|
||||
To train Llama 3.1 70B FP8 on 8 nodes, run:
|
||||
|
||||
@@ -879,20 +812,18 @@ to launch the multi-node workload. Use the following steps to setup your environ
|
||||
# Adjust the training parameters.
|
||||
# For example, `global_batch_size: 8 * #single_node_bs` for 8 nodes in this case.
|
||||
NNODES=8 \
|
||||
EXP=examples/megatron/configs/llama3.1_70B-pretrain.yaml \
|
||||
EXP=examples/megatron/configs/MI300X/llama3.1_70B-FP8-pretrain.yaml \
|
||||
bash examples/run_slurm_pretrain.sh \
|
||||
--micro_batch_size 4 \
|
||||
--global_batch_size 256 \
|
||||
--recompute_num_layers 80 \
|
||||
--no_fp8_weight_transpose_cache true \
|
||||
--fp8 hybrid
|
||||
|
||||
To train Llama 3.1 70B BF16 on 8 nodes, run:
|
||||
|
||||
.. code-block:: shell
|
||||
|
||||
NNODES=8 \
|
||||
EXP=examples/megatron/configs/llama3.1_70B-pretrain.yaml \
|
||||
EXP=examples/megatron/configs/MI300X/llama3.1_70B-BF16-pretrain.yaml \
|
||||
bash examples/run_slurm_pretrain.sh \
|
||||
--micro_batch_size 1 \
|
||||
--global_batch_size 256 \
|
||||
@@ -902,7 +833,7 @@ to launch the multi-node workload. Use the following steps to setup your environ
|
||||
|
||||
Once setup is complete, run the appropriate training command.
|
||||
The following run commands are tailored to Llama 2 70B.
|
||||
See :ref:`amd-primus-megatron-lm-model-support-v2510` to switch to another available model.
|
||||
See :ref:`amd-primus-megatron-lm-model-support-v25.11` to switch to another available model.
|
||||
|
||||
To train Llama 2 70B FP8 on 8 nodes, run:
|
||||
|
||||
@@ -911,20 +842,18 @@ to launch the multi-node workload. Use the following steps to setup your environ
|
||||
# Adjust the training parameters.
|
||||
# For example, `global_batch_size: 8 * #single_node_bs` for 8 nodes in this case.
|
||||
NNODES=8 \
|
||||
EXP=examples/megatron/configs/llama2_70B-pretrain.yaml \
|
||||
EXP=examples/megatron/configs/MI300X/llama2_70B-FP8-pretrain.yaml \
|
||||
bash examples/run_slurm_pretrain.sh \
|
||||
--micro_batch_size 10 \
|
||||
--global_batch_size 640 \
|
||||
--recompute_num_layers 80 \
|
||||
--no_fp8_weight_transpose_cache true \
|
||||
--fp8 hybrid
|
||||
|
||||
To train Llama 2 70B BF16 on 8 nodes, run:
|
||||
|
||||
.. code-block:: shell
|
||||
|
||||
NNODES=8 \
|
||||
EXP=examples/megatron/configs/llama2_70B-pretrain.yaml \
|
||||
EXP=examples/megatron/configs/MI300X/llama2_70B-BF16-pretrain.yaml \
|
||||
bash ./examples/run_slurm_pretrain.sh \
|
||||
--micro_batch_size 2 \
|
||||
--global_batch_size 1536 \
|
||||
@@ -934,7 +863,7 @@ to launch the multi-node workload. Use the following steps to setup your environ
|
||||
|
||||
Once setup is complete, run the appropriate training command.
|
||||
The following run commands are tailored to Llama 3.3 70B.
|
||||
See :ref:`amd-primus-megatron-lm-model-support-v2510` to switch to another available model.
|
||||
See :ref:`amd-primus-megatron-lm-model-support-v25.11` to switch to another available model.
|
||||
|
||||
To train Llama 3.3 70B FP8 on 8 nodes, run:
|
||||
|
||||
@@ -943,20 +872,18 @@ to launch the multi-node workload. Use the following steps to setup your environ
|
||||
# Adjust the training parameters.
|
||||
# For example, `global_batch_size: 8 * #single_node_bs` for 8 nodes in this case
|
||||
NNODES=8 \
|
||||
EXP=examples/megatron/configs/llama3.3_70B-pretrain.yaml \
|
||||
EXP=examples/megatron/configs/MI300X/llama3.3_70B-FP8-pretrain.yaml \
|
||||
bash examples/run_slurm_pretrain.sh \
|
||||
--micro_batch_size 4 \
|
||||
--global_batch_size 256 \
|
||||
--recompute_num_layers 80 \
|
||||
--no_fp8_weight_transpose_cache true \
|
||||
--fp8 hybrid
|
||||
|
||||
To train Llama 3.3 70B BF16 on 8 nodes, run:
|
||||
|
||||
.. code-block:: shell
|
||||
|
||||
NNODES=8 \
|
||||
EXP=examples/megatron/configs/llama3.3_70B-pretrain.yaml \
|
||||
EXP=examples/megatron/configs/MI300X/llama3.3_70B-BF16-pretrain.yaml \
|
||||
bash examples/run_slurm_pretrain.sh \
|
||||
--micro_batch_size 1 \
|
||||
--global_batch_size 256 \
|
||||
@@ -966,7 +893,7 @@ to launch the multi-node workload. Use the following steps to setup your environ
|
||||
|
||||
Once setup is complete, run the appropriate training command.
|
||||
The following run commands are tailored to Llama 2 70B.
|
||||
See :ref:`amd-primus-megatron-lm-model-support-v2510` to switch to another available model.
|
||||
See :ref:`amd-primus-megatron-lm-model-support-v25.11` to switch to another available model.
|
||||
|
||||
To train Mixtral 8x7B BF16 on 8 nodes, run:
|
||||
|
||||
@@ -975,7 +902,7 @@ to launch the multi-node workload. Use the following steps to setup your environ
|
||||
# Adjust the training parameters.
|
||||
# For example, `global_batch_size: 8 * #single_node_bs` for 8 nodes in this case
|
||||
NNODES=8 \
|
||||
EXP=examples/megatron/configs/mixtral_8x7B_v0.1-pretrain.yaml \
|
||||
EXP=examples/megatron/configs/MI300X/mixtral_8x7B_v0.1-BF16-pretrain.yaml \
|
||||
bash examples/run_slurm_pretrain.sh \
|
||||
--micro_batch_size 2 \
|
||||
--global_batch_size 256
|
||||
@@ -984,7 +911,7 @@ to launch the multi-node workload. Use the following steps to setup your environ
|
||||
|
||||
Once setup is complete, run the appropriate training command.
|
||||
The following run commands are tailored to Llama 2 70B.
|
||||
See :ref:`amd-primus-megatron-lm-model-support-v2510` to switch to another available model.
|
||||
See :ref:`amd-primus-megatron-lm-model-support-v25.11` to switch to another available model.
|
||||
|
||||
To train Qwen2.5 72B FP8 on 8 nodes, run:
|
||||
|
||||
@@ -993,15 +920,13 @@ to launch the multi-node workload. Use the following steps to setup your environ
|
||||
# Adjust the training parameters.
|
||||
# For example, `global_batch_size: 8 * #single_node_bs` for 8 nodes in this case
|
||||
NNODES=8 \
|
||||
EXP=examples/megatron/configs/qwen2.5_72B-pretrain.yaml \
|
||||
EXP=examples/megatron/configs/qwen2.5_72B-FP8-pretrain.yaml \
|
||||
bash examples/run_slurm_pretrain.sh \
|
||||
--micro_batch_size 8 \
|
||||
--global_batch_size 512 \
|
||||
--recompute_num_layers 80 \
|
||||
--no_fp8_weight_transpose_cache true \
|
||||
--fp8 hybrid
|
||||
|
||||
.. _amd-primus-megatron-lm-benchmark-test-vars-v2510:
|
||||
.. _amd-primus-megatron-lm-benchmark-test-vars-v25.11:
|
||||
|
||||
Key options
|
||||
-----------
|
||||
|
||||
@@ -45,7 +45,7 @@ with Primus Turbo optimizations.
|
||||
- {{ component_version }}
|
||||
{% endfor %}
|
||||
|
||||
.. _amd-primus-pytorch-model-support-v2510:
|
||||
.. _amd-primus-pytorch-model-support-v25.11:
|
||||
|
||||
Supported models
|
||||
================
|
||||
@@ -91,7 +91,7 @@ vary by model -- select one to get started.
|
||||
For additional workloads, including Llama 3.3, Llama 3.2, Llama 2, GPT OSS, Qwen, and Flux models,
|
||||
see the documentation :doc:`pytorch-training` (without Primus)
|
||||
|
||||
.. _amd-primus-pytorch-performance-measurements-v2510:
|
||||
.. _amd-primus-pytorch-performance-measurements-v25.11:
|
||||
|
||||
System validation
|
||||
=================
|
||||
@@ -146,7 +146,7 @@ tweak some configurations (such as batch sizes).
|
||||
.. container:: model-doc {{ model.mad_tag }}
|
||||
|
||||
The following run command is tailored to {{ model.model }}.
|
||||
See :ref:`amd-primus-pytorch-model-support-v2510` to switch to another available model.
|
||||
See :ref:`amd-primus-pytorch-model-support-v25.11` to switch to another available model.
|
||||
|
||||
1. Clone the ROCm Model Automation and Dashboarding (`<https://github.com/ROCm/MAD>`__) repository to a local
|
||||
directory and install the required packages on the host machine.
|
||||
@@ -184,7 +184,7 @@ tweak some configurations (such as batch sizes).
|
||||
.. container:: model-doc {{ model.mad_tag }}
|
||||
|
||||
The following run commands are tailored to {{ model.model }}.
|
||||
See :ref:`amd-primus-pytorch-model-support-v2510` to switch to another available model.
|
||||
See :ref:`amd-primus-pytorch-model-support-v25.11` to switch to another available model.
|
||||
|
||||
.. rubric:: Download the Docker image and required packages
|
||||
|
||||
@@ -220,6 +220,9 @@ tweak some configurations (such as batch sizes).
|
||||
docker start training_env
|
||||
docker exec -it training_env bash
|
||||
|
||||
The Docker container hosts verified commit ``c4c083de`` of the `Primus
|
||||
<https://github.com/AMD-AGI/Primus/tree/c4c083de64ba3e8f19ccc9629411267108931f9e/>`__ repository.
|
||||
|
||||
.. rubric:: Prepare training datasets and dependencies
|
||||
|
||||
The following benchmarking examples require downloading models and datasets
|
||||
@@ -255,7 +258,7 @@ tweak some configurations (such as batch sizes).
|
||||
.. code-block:: shell
|
||||
|
||||
EXP=examples/torchtitan/configs/MI355X/llama3.1_8B-BF16-pretrain.yaml \
|
||||
bash examples/run_pretrain.sh --training.local_batch_size 6
|
||||
bash examples/run_pretrain.sh
|
||||
|
||||
.. tab-item:: MI325X
|
||||
:sync: MI325X
|
||||
@@ -263,7 +266,7 @@ tweak some configurations (such as batch sizes).
|
||||
.. code-block:: shell
|
||||
|
||||
EXP=examples/torchtitan/configs/MI300X/llama3.1_8B-BF16-pretrain.yaml \
|
||||
bash examples/run_pretrain.sh --training.local_batch_size 6
|
||||
bash examples/run_pretrain.sh --training.local_batch_size 6
|
||||
|
||||
.. tab-item:: MI300X
|
||||
:sync: MI300X
|
||||
@@ -271,8 +274,7 @@ tweak some configurations (such as batch sizes).
|
||||
.. code-block:: shell
|
||||
|
||||
EXP=examples/torchtitan/configs/MI300X/llama3.1_8B-BF16-pretrain.yaml \
|
||||
bash examples/run_pretrain.sh --training.local_batch_size 4
|
||||
|
||||
bash examples/run_pretrain.sh
|
||||
|
||||
To train Llama 3.1 8B with FP8 precision, use the following command.
|
||||
|
||||
@@ -284,7 +286,7 @@ tweak some configurations (such as batch sizes).
|
||||
.. code-block:: shell
|
||||
|
||||
EXP=examples/torchtitan/configs/MI355X/llama3.1_8B-BF16-pretrain.yaml \
|
||||
bash examples/run_pretrain.sh --training.local_batch_size 8
|
||||
bash examples/run_pretrain.sh
|
||||
|
||||
.. tab-item:: MI325X
|
||||
:sync: MI325X
|
||||
@@ -292,7 +294,7 @@ tweak some configurations (such as batch sizes).
|
||||
.. code-block:: shell
|
||||
|
||||
EXP=examples/torchtitan/configs/MI300X/llama3.1_8B-FP8-pretrain.yaml \
|
||||
bash examples/run_pretrain.sh --training.local_batch_size 7
|
||||
bash examples/run_pretrain.sh --training.local_batch_size 7
|
||||
|
||||
.. tab-item:: MI300X
|
||||
:sync: MI300X
|
||||
@@ -300,7 +302,7 @@ tweak some configurations (such as batch sizes).
|
||||
.. code-block:: shell
|
||||
|
||||
EXP=examples/torchtitan/configs/MI300X/llama3.1_8B-FP8-pretrain.yaml \
|
||||
bash examples/run_pretrain.sh --training.local_batch_size 5
|
||||
bash examples/run_pretrain.sh
|
||||
|
||||
.. container:: model-doc primus_pyt_train_llama-3.1-70b
|
||||
|
||||
@@ -314,7 +316,7 @@ tweak some configurations (such as batch sizes).
|
||||
.. code-block:: shell
|
||||
|
||||
EXP=examples/torchtitan/configs/MI355X/llama3.1_70B-BF16-pretrain.yaml \
|
||||
bash examples/run_pretrain.sh --training.local_batch_size 8
|
||||
bash examples/run_pretrain.sh
|
||||
|
||||
.. tab-item:: MI325X
|
||||
:sync: MI325X
|
||||
@@ -322,7 +324,7 @@ tweak some configurations (such as batch sizes).
|
||||
.. code-block:: shell
|
||||
|
||||
EXP=examples/torchtitan/configs/MI300X/llama3.1_70B-BF16-pretrain.yaml \
|
||||
bash examples/run_pretrain.sh --training.local_batch_size 6
|
||||
bash examples/run_pretrain.sh --training.local_batch_size 6
|
||||
|
||||
.. tab-item:: MI300X
|
||||
:sync: MI300X
|
||||
@@ -330,7 +332,7 @@ tweak some configurations (such as batch sizes).
|
||||
.. code-block:: shell
|
||||
|
||||
EXP=examples/torchtitan/configs/MI300X/llama3.1_70B-BF16-pretrain.yaml \
|
||||
bash examples/run_pretrain.sh --training.local_batch_size 4
|
||||
bash examples/run_pretrain.sh
|
||||
|
||||
To train Llama 3.1 70B with FP8 precision, use the following command.
|
||||
|
||||
@@ -342,7 +344,7 @@ tweak some configurations (such as batch sizes).
|
||||
.. code-block:: shell
|
||||
|
||||
EXP=examples/torchtitan/configs/MI355X/llama3.1_70B-FP8-pretrain.yaml \
|
||||
bash examples/run_pretrain.sh --training.local_batch_size 6
|
||||
bash examples/run_pretrain.sh
|
||||
|
||||
.. tab-item:: MI325X
|
||||
:sync: MI325X
|
||||
@@ -350,7 +352,7 @@ tweak some configurations (such as batch sizes).
|
||||
.. code-block:: shell
|
||||
|
||||
EXP=examples/torchtitan/configs/MI300X/llama3.1_70B-FP8-pretrain.yaml \
|
||||
bash examples/run_pretrain.sh --training.local_batch_size 5
|
||||
bash examples/run_pretrain.sh --training.local_batch_size 5
|
||||
|
||||
.. tab-item:: MI300X
|
||||
:sync: MI300X
|
||||
@@ -358,11 +360,11 @@ tweak some configurations (such as batch sizes).
|
||||
.. code-block:: shell
|
||||
|
||||
EXP=examples/torchtitan/configs/MI300X/llama3.1_70B-FP8-pretrain.yaml \
|
||||
bash examples/run_pretrain.sh --training.local_batch_size 3
|
||||
bash examples/run_pretrain.sh
|
||||
|
||||
.. container:: model-doc primus_pyt_train_deepseek-v2
|
||||
.. container:: model-doc primus_pyt_train_deepseek-v3-16b
|
||||
|
||||
Use the following command to run train DeepSeek V2 16B with BF16 precision using Primus torchtitan.
|
||||
Use the following command to run train DeepSeek V3 16B with BF16 precision using Primus torchtitan.
|
||||
|
||||
.. tab-set::
|
||||
|
||||
@@ -372,7 +374,7 @@ tweak some configurations (such as batch sizes).
|
||||
.. code-block:: shell
|
||||
|
||||
EXP=examples/torchtitan/configs/MI355X/deepseek_v3_16b-pretrain.yaml \
|
||||
bash examples/run_pretrain.sh --training.local_batch_size 16
|
||||
bash examples/run_pretrain.sh
|
||||
|
||||
.. tab-item:: MI325X
|
||||
:sync: MI325X
|
||||
@@ -380,7 +382,7 @@ tweak some configurations (such as batch sizes).
|
||||
.. code-block:: shell
|
||||
|
||||
EXP=examples/torchtitan/configs/MI300X/deepseek_v3_16b-pretrain.yaml \
|
||||
bash examples/run_pretrain.sh --training.local_batch_size 10
|
||||
bash examples/run_pretrain.sh --training.local_batch_size 10
|
||||
|
||||
.. tab-item:: MI300X
|
||||
:sync: MI300X
|
||||
@@ -388,35 +390,7 @@ tweak some configurations (such as batch sizes).
|
||||
.. code-block:: shell
|
||||
|
||||
EXP=examples/torchtitan/configs/MI300X/deepseek_v3_16b-pretrain.yaml \
|
||||
bash examples/run_pretrain.sh --training.local_batch_size 8
|
||||
|
||||
To train DeepSeek V2 16B with FP8 precision, use the following command.
|
||||
|
||||
.. tab-set::
|
||||
|
||||
.. tab-item:: MI355X and MI350X
|
||||
:sync: MI355X
|
||||
|
||||
.. code-block:: shell
|
||||
|
||||
EXP=examples/torchtitan/configs/MI355X/deepseek_v3_16b-pretrain.yaml \
|
||||
bash examples/run_pretrain.sh --training.local_batch_size 16
|
||||
|
||||
.. tab-item:: MI325X
|
||||
:sync: MI325X
|
||||
|
||||
.. code-block:: shell
|
||||
|
||||
EXP=examples/torchtitan/configs/MI300X/deepseek_v3_16b-pretrain.yaml \
|
||||
bash examples/run_pretrain.sh --training.local_batch_size 8
|
||||
|
||||
.. tab-item:: MI300X
|
||||
:sync: MI300X
|
||||
|
||||
.. code-block:: shell
|
||||
|
||||
EXP=examples/torchtitan/configs/MI300X/deepseek_v3_16b-pretrain.yaml \
|
||||
bash examples/run_pretrain.sh --training.local_batch_size 8
|
||||
bash examples/run_pretrain.sh
|
||||
{% endfor %}
|
||||
{% endfor %}
|
||||
|
||||
|
||||
@@ -43,7 +43,7 @@ training workloads:
|
||||
- {{ component_version }}
|
||||
{% endfor %}
|
||||
|
||||
.. _amd-pytorch-training-model-support-v2510:
|
||||
.. _amd-pytorch-training-model-support-v25.11:
|
||||
|
||||
Supported models
|
||||
================
|
||||
@@ -85,7 +85,7 @@ one to get started.
|
||||
</div>
|
||||
</div>
|
||||
|
||||
.. _amd-pytorch-training-supported-training-modes-v2510:
|
||||
.. _amd-pytorch-training-supported-training-modes-v25.11:
|
||||
|
||||
The following table lists supported training modes per model.
|
||||
|
||||
@@ -120,7 +120,7 @@ The following table lists supported training modes per model.
|
||||
unlisted fine-tuning methods by using an existing file in the
|
||||
``/workspace/torchtune/recipes/configs`` directory as a template.
|
||||
|
||||
.. _amd-pytorch-training-performance-measurements-v2510:
|
||||
.. _amd-pytorch-training-performance-measurements-v25.11:
|
||||
|
||||
Performance measurements
|
||||
========================
|
||||
@@ -176,7 +176,7 @@ Run training
|
||||
.. container:: model-doc {{ model.mad_tag }}
|
||||
|
||||
The following run command is tailored to {{ model.model }}.
|
||||
See :ref:`amd-pytorch-training-model-support-v2510` to switch to another available model.
|
||||
See :ref:`amd-pytorch-training-model-support-v25.11` to switch to another available model.
|
||||
|
||||
1. Clone the ROCm Model Automation and Dashboarding (`<https://github.com/ROCm/MAD>`__) repository to a local
|
||||
directory and install the required packages on the host machine.
|
||||
@@ -214,7 +214,7 @@ Run training
|
||||
.. container:: model-doc {{ model.mad_tag }}
|
||||
|
||||
The following commands are tailored to {{ model.model }}.
|
||||
See :ref:`amd-pytorch-training-model-support-v2510` to switch to another available model.
|
||||
See :ref:`amd-pytorch-training-model-support-v25.11` to switch to another available model.
|
||||
|
||||
{% endfor %}
|
||||
{% endfor %}
|
||||
@@ -532,7 +532,7 @@ Run training
|
||||
|
||||
To start the fine-tuning benchmark, use the following command with the
|
||||
appropriate options. See the following list of options and their descriptions.
|
||||
See :ref:`supported training modes <amd-pytorch-training-supported-training-modes-v2510>`.
|
||||
See :ref:`supported training modes <amd-pytorch-training-supported-training-modes-v25.11>`.
|
||||
|
||||
.. code-block:: shell
|
||||
|
||||
@@ -597,7 +597,7 @@ Run training
|
||||
|
||||
For examples of benchmarking commands, see `<https://github.com/ROCm/MAD/tree/develop/benchmark/pytorch_train#benchmarking-examples>`__.
|
||||
|
||||
.. _amd-pytorch-training-multinode-examples-v2510:
|
||||
.. _amd-pytorch-training-multinode-examples-v25.11:
|
||||
|
||||
Multi-node training
|
||||
-------------------
|
||||
|
||||
@@ -75,8 +75,14 @@ subtrees:
|
||||
- entries:
|
||||
- file: how-to/rocm-for-ai/training/benchmark-docker/primus-megatron.rst
|
||||
title: Train a model with Primus and Megatron-LM
|
||||
entries:
|
||||
- file: how-to/rocm-for-ai/training/benchmark-docker/megatron-lm.rst
|
||||
title: Train a model with Megatron-LM
|
||||
- file: how-to/rocm-for-ai/training/benchmark-docker/primus-pytorch.rst
|
||||
title: Train a model with Primus and PyTorch
|
||||
entries:
|
||||
- file: how-to/rocm-for-ai/training/benchmark-docker/pytorch-training.rst
|
||||
title: Train a model with PyTorch
|
||||
- file: how-to/rocm-for-ai/training/benchmark-docker/jax-maxtext.rst
|
||||
title: Train a model with JAX MaxText
|
||||
- file: how-to/rocm-for-ai/training/benchmark-docker/mpt-llm-foundry
|
||||
|
||||
Reference in New Issue
Block a user