Skip to content

Megatron + DeepSpeed + Pipeline Parallelism#1

Merged
jeffra merged 15 commits intomainfrom
megatron-2.4-ds-pipe
Jul 2, 2021
Merged

Megatron + DeepSpeed + Pipeline Parallelism#1
jeffra merged 15 commits intomainfrom
megatron-2.4-ds-pipe

Conversation

@jeffra
Copy link
Copy Markdown

@jeffra jeffra commented Jun 23, 2021

No description provided.

@jeffra jeffra changed the title Megatron 2.4 ds pipe Megatron DeepSpeed + Pipeline Parallelism Jun 23, 2021
@jeffra jeffra changed the title Megatron DeepSpeed + Pipeline Parallelism Megatron + DeepSpeed + Pipeline Parallelism Jun 23, 2021
@jeffra jeffra merged commit b56b50b into main Jul 2, 2021
jeffra pushed a commit that referenced this pull request Jan 18, 2022
wuhuachaocoding pushed a commit to wuhuachaocoding/Megatron-DeepSpeed that referenced this pull request Aug 15, 2022
* Enable Megatron-LM workload on ROCm (deepspeedai#1)

* Enable Megatron workload on ROCm

* Added ds_pretrain_gpt_350M_dense_pipeclean.sh

* removed a file

* Removed an extra line

* Fix to resolve the below rsqrtf() error on ROCm

/root/Megatron-DeepSpeed/megatron/fused_kernels/layer_norm_hip_kernel.hip:298:10: error: no matching function for call to 'rsqrtf'
  return rsqrtf(v);
         ^~~~~~
/opt/rocm-5.2.0/llvm/lib/clang/14.0.0/include/__clang_hip_math.h:521:7: note: candidate function not viable: call to __device__ function from __host__ function
float rsqrtf(float __x) { return __ocml_rsqrt_f32(__x); }
      ^

* Simplified code

* Simplified the code

* Removed extra spaces
yueming-yuan referenced this pull request in Supercomputing-System-AI-Lab/Megatron-DeepSpeed Sep 14, 2025
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants