Skip to content
This repository was archived by the owner on Jan 15, 2026. It is now read-only.

Mixtral 7B yaml file with mcore distributed optim #328

Draft
gdengk wants to merge 1 commit into
base: main
Choose a base branch
from
Draft

Mixtral 7B yaml file with mcore distributed optim #328

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
10 changes: 6 additions & 4 deletions launcher_scripts/conf/training/mixtral/mixtral_8x7b.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -52,9 +52,9 @@ model:
micro_batch_size: 1
global_batch_size: 256
rampup_batch_size: null
tensor_model_parallel_size: 8
pipeline_model_parallel_size: 4
expert_model_parallel_size: 1
tensor_model_parallel_size: 2
pipeline_model_parallel_size: 1
expert_model_parallel_size: 8
virtual_pipeline_model_parallel_size: null
encoder_seq_length: 4096
max_position_embeddings: 32768
Expand Down Expand Up @@ -145,7 +145,9 @@ model:
- 0
gen_shape: false
optim:
name: distributed_fused_adam
name: mcore_distributed_optim
overlap_grad_sync: true
overlap_param_sync: true
lr: 0.0001
weight_decay: 0.1
betas:
Expand Down