Files
Hy-MT2-7B-GGUF/train/deepspeed_support/ds_zero2_offload.json
ModelHub XC 5212a98298 初始化项目,由ModelHub XC社区提供模型
Model: Tencent-Hunyuan/Hy-MT2-7B-GGUF
Source: Original Platform
2026-08-08 21:25:13 +08:00

28 lines
676 B
JSON

{
"fp16": {
"enabled": false
},
"bf16": {
"enabled": true
},
"zero_optimization": {
"stage": 2,
"offload_optimizer": {
"device": "cpu",
"pin_memory": true
},
"allgather_partitions": true,
"allgather_bucket_size": 1e8,
"overlap_comm": true,
"reduce_scatter": true,
"reduce_bucket_size": 1e8,
"contiguous_gradients": true
},
"gradient_accumulation_steps": "auto",
"gradient_clipping": "auto",
"steps_per_print": 10,
"train_batch_size": "auto",
"train_micro_batch_size_per_gpu": "auto",
"wall_clock_breakdown": false
}