hailin
/
jd_train
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

.

This commit is contained in:
hailin 2025-08-29 23:36:05 +08:00
parent f8262571ab
commit 1dd1a1f6c5
1 changed files with 11 additions and 5 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

14
ds_config_zero3.json
View File

@ -1,6 +1,7 @@
{
"train_micro_batch_size_per_gpu": "auto",
"gradient_accumulation_steps": "auto",
"train_micro_batch_size_per_gpu": 1,
"gradient_accumulation_steps": 1,
"zero_optimization": {
"stage": 3,
"overlap_comm": true,
@ -11,12 +12,17 @@
"offload_optimizer": { "device": "cpu", "pin_memory": true },
"offload_param": { "device": "none" }
},
"fp16": { "enabled": false },
"bf16": { "enabled": true },
"fp16": { "enabled": false },
"optimizer": {
"type": "CPUAdam",
"params": { "lr": 2e-5, "betas": [0.9, 0.999], "eps": 1e-8, "weight_decay": 0.1 }
},
"gradient_clipping": 1.0,
"zero_force_ds_cpu_optimizer": true
"zero_force_ds_cpu_optimizer": true,
"wall_clock_breakdown": false
}