Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

update gpt&baichuan&qwen ce name #9697

Merged
merged 5 commits into from
Dec 26, 2024
Merged

update gpt&baichuan&qwen ce name #9697

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
17f018c
update gpt&baichuan&qwen ce name
blacksheep-Aristotle Dec 25, 2024
1e63093
update gpt&baichuan&qwen ce name
blacksheep-Aristotle Dec 25, 2024
f03e016
update gpt&baichuan&qwen ce name
blacksheep-Aristotle Dec 25, 2024
fc39c35
update gpt&baichuan&qwen ce name
blacksheep-Aristotle Dec 25, 2024
658206d
update gpt&baichuan&qwen ce name
blacksheep-Aristotle Dec 25, 2024
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
2 changes: 1 addition & 1 deletion ...bf16_DP1_MP2_PP4_VPP1_Sharding4_Stage1.sh → ...bf16_DP1_MP2_PP4_VPP5_Sharding4_Stage1.sh
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -12,7 +12,7 @@
# See the License for the specific language governing permissions and
# limitations under the License.

param="model_item=gpt3-13b_pretrain "
param="model_item=gpt-3-13b_pretrain "
param+="run_mode=DP1_MP2_PP4_VPP5_Sharding4_Stage1 "
param+="device_num=N4C32 "
param+="global_batch_size=128 "
Expand Down
4 changes: 2 additions & 2 deletions ...bf16_DP1_MP4_PP2_1F1B_Sharding4_Stage1.sh → ...bf16_DP1_MP4_PP2_1F1B_Sharding4_Stage1.sh
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -12,8 +12,8 @@
# See the License for the specific language governing permissions and
# limitations under the License.

param="model_item=baichuan-inc-baichaun-2-13b_pretrain "
param+="run_mode=DP1_MP2_PP4_1F1B_Sharding8_Stage2 "
param="model_item=baichuan-inc-baichuan-2-13b_pretrain "
param+="run_mode=DP1_MP4_PP2_1F1B_Sharding4_Stage1 "
param+="device_num=N4C32 "
param+="global_batch_size=128 "
param+="nnodes=4 "
Expand Down
4 changes: 2 additions & 2 deletions .../static/auto_parallel/baichuan2/pretrain_config_baichuan2_13b/pretrain-baichuan2_13b.json
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -44,12 +44,12 @@
"fused_linear": 1,
"fused_linear_param_grad_add": 1,
"use_fused_rope": true,
"use_fused_rms_norm": false,
"use_fused_rms_norm": true,
"max_seq_length": 4096,
"sequence_parallel": false,
"sharding": "stage1",
"sharding_parallel_config": "enable_stage1_tensor_fusion enable_stage1_overlap",
"tensor_parallel_config": "enable_mp_async_allreduce",
"tensor_parallel_config": "enable_mp_async_allreduce replace_with_parallel_cross_entropy",
"data_parallel_config": "enable_allreduce_avg_in_gradinent_scale gradient_sync_after_accumulate",
"pipeline_parallel_config": "enable_send_recv_overlap enable_split_backward"
}
2 changes: 1 addition & 1 deletion ...bf16_DP1_MP1_PP4_1F1B_Sharding4_Stage1.sh → ...bf16_DP1_MP2_PP4_1F1B_Sharding4_Stage1.sh
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -12,7 +12,7 @@
# See the License for the specific language governing permissions and
# limitations under the License.

param="model_item=gpt3-13b_pretrain_dy2st "
param="model_item=gpt-3-13b_pretrain_dy2st "
param+="run_mode=DP1_MP2_PP4_1F1B_Sharding4_Stage1 "
param+="device_num=N4C32 "
param+="global_batch_size=128 "
Expand Down
4 changes: 2 additions & 2 deletions tests/test_tipc/static/auto_parallel/gpt3/pretrain_config_gpt3_13b/pretrain-gpt3_13b.json
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -39,7 +39,7 @@
"fuse_attention_qkv": 1,
"fused_linear_param_grad_add": 1,
"use_fused_rope": true,
"use_fused_rms_norm": false,
"use_fused_rms_norm": true,
"recompute": 0,
"recompute_use_reentrant": true,
"recompute_granularity": "full",
Expand All @@ -52,7 +52,7 @@
"attention_probs_dropout_prob": 0.1,
"hidden_dropout_prob": 0.1,
"sharding_parallel_config": "enable_stage1_tensor_fusion enable_stage1_overlap",
"tensor_parallel_config": "enable_mp_async_allreduce",
"tensor_parallel_config": "enable_mp_async_allreduce replace_with_parallel_cross_entropy",
"data_parallel_config": "enable_allreduce_avg_in_gradinent_scale gradient_sync_after_accumulate",
"pipeline_parallel_config": "enable_send_recv_overlap enable_split_backward"
}
2 changes: 1 addition & 1 deletion ...test_tipc/static/auto_parallel/llama2/pretrain_config_llama2_70b/pretrain-llama2_70b.json
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -53,7 +53,7 @@
"pipeline_schedule_mode": "VPP",
"data_parallel_config": "enable_allreduce_avg_in_gradinent_scale gradient_sync_after_accumulate",
"sharding_parallel_config": "enable_stage1_overlap",
"tensor_parallel_config": "enable_mp_async_allreduce",
"tensor_parallel_config": "enable_mp_async_allreduce replace_with_parallel_cross_entropy",
"max_seq_length": 4096,
"to_static": true,
"eliminate_transpose": 1,
Expand Down
2 changes: 1 addition & 1 deletion ...rallel/qwen/N4C32/qwen-14b_pretrain_dy2st_bs128_bf16_DP1_MP2_PP4_1F1B_Sharding4_Stage1.sh
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -12,7 +12,7 @@
# See the License for the specific language governing permissions and
# limitations under the License.

param="model_item=qwen-2-14b_pretrain_dy2st "
param="model_item=qwen-14b_pretrain_dy2st "
param+="run_mode=DP1_MP2_PP4_1F1B_Sharding4_Stage1 "
param+="device_num=N4C32 "
param+="global_batch_size=128 "
Expand Down
2 changes: 1 addition & 1 deletion tests/test_tipc/static/auto_parallel/qwen/pretrain_config_qwen_14b/pretrain-qwen_14b.json
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -50,6 +50,6 @@
"auto_parallel_resume_form_hybrid_parallel": true,
"data_parallel_config": "enable_allreduce_avg_in_gradinent_scale gradient_sync_after_accumulate",
"sharding_parallel_config": "enable_stage1_overlap",
"tensor_parallel_config": "enable_mp_async_allreduce",
"tensor_parallel_config": "enable_mp_async_allreduce replace_with_parallel_cross_entropy",
"pipeline_parallel_config": "enable_send_recv_overlap enable_split_backward"
}