github/tinygrad
1
1
Fork 0
mirror of https://github.com/tinygrad/tinygrad.git synced 2026-01-08 22:48:25 -05:00
Code Issues Packages Projects Releases Wiki Activity

update bert mi300x script (#9872)

Browse Source 
2 runs failed to converge in 10 back to back runs, increase total train steps and some beam params (2% faster step)
This commit is contained in:
chenyu
2025-04-13 10:07:36 -04:00
committed by GitHub
parent e201bc3e93
commit e2a40fb523
3 changed files with 5 additions and 5 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
examples/mlperf/training_submission_v5.0/tinycorp/benchmarks/bert/implementations/tinybox_8xMI300X/dev_beam.sh
View File

@@ -5,7 +5,7 @@ export MODEL="bert"
export DEFAULT_FLOAT="HALF" GPUS=8 BS=1024 EVAL_BS=1024
export OPT_BASE_LEARNING_RATE=0.0011 OPT_LAMB_BETA_1=0.60466 OPT_LAMB_BETA_2=0.85437 DECAY=0.1
export BEAM=3 BEAM_UOPS_MAX=4000 BEAM_UPCAST_MAX=256 BEAM_LOCAL_MAX=1024 BEAM_MIN_PROGRESS=5
export BEAM=3 BEAM_UOPS_MAX=6000 BEAM_UPCAST_MAX=256 BEAM_LOCAL_MAX=1024 BEAM_MIN_PROGRESS=10
export IGNORE_JIT_FIRST_BEAM=1 FREE_INTERMEDIATE=0
export BASEDIR="/raid/datasets/wiki"

4
examples/mlperf/training_submission_v5.0/tinycorp/benchmarks/bert/implementations/tinybox_8xMI300X/dev_run.sh
View File

@@ -6,9 +6,9 @@ export DEFAULT_FLOAT="HALF" GPUS=8 BS=1024 EVAL_BS=1024
# similar to https://github.com/mlcommons/training_results_v3.1/blob/d06288b2bd675a9d88e0e6181f5bb5626b71ec19/Quanta_Cloud_Technology/results/D54U-3U/bert/result_1.txt#L54
export OPT_BASE_LEARNING_RATE=0.0011 OPT_LAMB_BETA_1=0.60466 OPT_LAMB_BETA_2=0.85437 DECAY=0.1
export TRAIN_STEPS=3600
export TRAIN_STEPS=3900
export BEAM=3 BEAM_UOPS_MAX=4000 BEAM_UPCAST_MAX=256 BEAM_LOCAL_MAX=1024 BEAM_MIN_PROGRESS=5
export BEAM=3 BEAM_UOPS_MAX=6000 BEAM_UPCAST_MAX=256 BEAM_LOCAL_MAX=1024 BEAM_MIN_PROGRESS=10
export IGNORE_JIT_FIRST_BEAM=1 FREE_INTERMEDIATE=0
export BASEDIR="/raid/datasets/wiki"

4
examples/mlperf/training_submission_v5.0/tinycorp/benchmarks/bert/implementations/tinybox_8xMI300X/run_and_time.sh
View File

@@ -7,9 +7,9 @@ export DEFAULT_FLOAT="HALF" GPUS=8 BS=1024 EVAL_BS=1024
# similar to https://github.com/mlcommons/training_results_v3.1/blob/d06288b2bd675a9d88e0e6181f5bb5626b71ec19/Quanta_Cloud_Technology/results/D54U-3U/bert/result_1.txt#L54
export OPT_BASE_LEARNING_RATE=0.0011 OPT_LAMB_BETA_1=0.60466 OPT_LAMB_BETA_2=0.85437 DECAY=0.1
export TRAIN_STEPS=3600
export TRAIN_STEPS=3900
export BEAM=3 BEAM_UOPS_MAX=4000 BEAM_UPCAST_MAX=256 BEAM_LOCAL_MAX=1024 BEAM_MIN_PROGRESS=5
export BEAM=3 BEAM_UOPS_MAX=6000 BEAM_UPCAST_MAX=256 BEAM_LOCAL_MAX=1024 BEAM_MIN_PROGRESS=10
export IGNORE_JIT_FIRST_BEAM=1 FREE_INTERMEDIATE=0
export BASEDIR="/raid/datasets/wiki"