Upload llama-1B/8_GPUS/dp-8_tp-1_pp-1_mbz-1

.gitattributes

@@ -128,3 +128,4 @@ llama-1B/8_GPUS/dp-2_tp-1_pp-4_mbz-2/profiler/ip-26-0-163-220_436335.17200509603
 llama-1B/8_GPUS/dp-8_tp-1_pp-1_mbz-2/profiler/ip-26-0-169-86_2283054.1720051611565005497.pt.trace.json filter=lfs diff=lfs merge=lfs -text
 llama-1B/8_GPUS/dp-4_tp-1_pp-2_mbz-2/profiler/ip-26-0-169-239_2659139.1720051804197315742.pt.trace.json filter=lfs diff=lfs merge=lfs -text
 llama-1B/8_GPUS/dp-1_tp-8_pp-1_mbz-16/profiler/ip-26-0-171-88_1107214.1720050881924334797.pt.trace.json filter=lfs diff=lfs merge=lfs -text
+llama-1B/8_GPUS/dp-8_tp-1_pp-1_mbz-1/profiler/ip-26-0-171-62_4129089.1720072629336630088.pt.trace.json filter=lfs diff=lfs merge=lfs -text

llama-1B/8_GPUS/dp-8_tp-1_pp-1_mbz-1/bench.slurm

@@ -5,7 +5,7 @@
 #SBATCH --partition=hopper-prod
 #SBATCH --nodes=1
 #SBATCH --gres=gpu:8
-#SBATCH --qos=normal
+#SBATCH --qos=low
 #SBATCH --ntasks-per-node=1
 #SBATCH --cpus-per-task=96
 #SBATCH --exclusive

llama-1B/8_GPUS/dp-8_tp-1_pp-1_mbz-1/log_metrics.csv

@@ -0,0 +1,21 @@
+iteration,consumed_tokens,elapsed_time_per_iteration_ms,tokens_per_sec,tokens_per_sec_per_gpu,global_batch_size,lm_loss,lr,model_tflops_per_gpu,hardware_tflops_per_gpu,grad_norm,memory_usage_MiB,peak_allocated_MiB,peak_reserved_MiB
+1,4190000.0000000005,23800.0,177000.0,22100.0,1020.0,11.3,0.0001,200.0,200.0,33.1,8040.89,16569.84,19382.0
+2,8390000.0,18200.0,230000.0,28800.0,1020.0,11.3,9.53e-05,261.0,261.0,33.3,8040.89,16569.84,19382.0
+3,12600000.0,18900.0,221000.0,27700.0,1020.0,16.0,9.05e-05,251.0,251.0,249.0,8040.89,16569.84,19382.0
+4,16800000.0,23400.0,179000.0,22400.0,1020.0,15.1,8.58e-05,203.0,203.0,41.6,8040.83,12538.47,19382.0
+5,21000000.0,23500.0,178000.0,22300.0,1020.0,10.8,8.11e-05,202.0,202.0,25.9,8040.83,16569.84,19382.0
+6,25200000.0,23700.0,177000.0,22100.0,1020.0,10.8,7.63e-05,201.0,201.0,18.9,8040.83,16569.84,19382.0
+7,29400000.0,19700.0,213000.0,26600.0,1020.0,10.2,7.16e-05,242.0,242.0,7.97,8040.83,16569.84,19382.0
+8,33600000.0,19200.0,219000.0,27400.0,1020.0,9.15,6.68e-05,248.0,248.0,6.46,8040.83,16569.84,19382.0
+9,37700000.0,19000.0,221000.0,27600.0,1020.0,11.2,6.21e-05,250.0,250.0,59.7,8040.83,16569.84,19382.0
+10,41900000.0,19000.0,221000.0,27600.0,1020.0,9.59,5.74e-05,250.0,250.0,44.0,8040.83,16569.84,19382.0
+11,46100000.0,19000.0,221000.0,27700.0,1020.0,8.08,5.26e-05,251.0,251.0,8.35,8040.83,16569.84,19382.0
+12,50300000.0,19000.0,221000.0,27700.0,1020.0,7.85,4.79e-05,251.0,251.0,5.09,8040.83,16569.84,19382.0
+13,54500000.0,19000.0,220000.0,27500.0,1020.0,7.7,4.32e-05,250.0,250.0,4.7,8040.83,16569.84,19382.0
+14,58700000.0,19000.0,220000.0,27500.0,1020.0,7.56,3.84e-05,250.0,250.0,5.17,8040.83,16569.84,19382.0
+15,62900000.0,19100.0,220000.0,27500.0,1020.0,7.4,3.37e-05,250.0,250.0,5.17,8040.83,16569.84,19382.0
+16,67099999.99999999,19000.0,221000.0,27600.0,1020.0,7.3,2.89e-05,250.0,250.0,5.34,8040.83,16569.84,19382.0
+17,71300000.0,19000.0,220000.0,27500.0,1020.0,7.22,2.42e-05,250.0,250.0,5.13,8040.83,16569.84,19382.0
+18,75500000.0,19000.0,220000.0,27500.0,1020.0,7.15,1.95e-05,250.0,250.0,5.05,8040.83,16569.84,19382.0
+19,79700000.0,19000.0,220000.0,27600.0,1020.0,7.08,1.47e-05,250.0,250.0,3.89,8040.83,16569.84,19382.0
+20,83900000.0,19000.0,220000.0,27500.0,1020.0,7.03,1e-05,250.0,250.0,2.9,,,

llama-1B/8_GPUS/dp-8_tp-1_pp-1_mbz-1/profiler.csv

@@ -0,0 +1,2 @@
+forward,backward
+0ms 892μs,0ms 314μs

llama-1B/8_GPUS/dp-8_tp-1_pp-1_mbz-1/profiler/ip-26-0-171-62_4129089.1720072629336630088.pt.trace.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a6c1bd4f234bf4644eb5736002da74d0a5708c2766efbd1f9b577de58aeb398c
+size 17204720966

llama-1B/8_GPUS/dp-8_tp-1_pp-1_mbz-1/status.txt

@@ -1 +1 @@
-fail
+completed