Hugging Face's logo

nanotron
/
old_bench

Model card Files Community
3outeille HF staff commited on
Commit
86c9c46
•
1 Parent(s): 710fa25

Upload llama-1B/64_GPUS/dp-1_tp-1_pp-64_mbz-16

Browse files
Files changed (2) hide show
  1. llama-1B/64_GPUS/dp-1_tp-1_pp-64_mbz-16/config.yaml +1 -1
  2. llama-1B/64_GPUS/dp-1_tp-1_pp-64_mbz-16/log.out +0 -0
llama-1B/64_GPUS/dp-1_tp-1_pp-64_mbz-16/config.yaml CHANGED
@@ -48,7 +48,7 @@ parallelism:
48
  dp: 1
49
  expert_parallel_size: 1
50
  pp: 64
51
- pp_engine: afab
52
  tp: 1
53
  tp_linear_async_communication: false
54
  tp_mode: REDUCE_SCATTER
 
48
  dp: 1
49
  expert_parallel_size: 1
50
  pp: 64
51
+ pp_engine: 1f1b
52
  tp: 1
53
  tp_linear_async_communication: false
54
  tp_mode: REDUCE_SCATTER
llama-1B/64_GPUS/dp-1_tp-1_pp-64_mbz-16/log.out CHANGED
The diff for this file is too large to render. See raw diff