Files
axolotl/deepspeed_configs/zero3.json
Wing Lian e207762928 fix deprecate deepspeed stage3_gather_16bit_weights_on_model_save arg (#2956) [skip ci]
* fix deprecate deepspeed stage3_gather_16bit_weights_on_model_save arg

* replace the rest of the migrated deepspeed params
2025-07-21 11:41:31 -04:00

32 lines
763 B
JSON

{
"zero_optimization": {
"stage": 3,
"overlap_comm": true,
"contiguous_gradients": true,
"sub_group_size": 0,
"reduce_bucket_size": "auto",
"prefetch_bucket_size": "auto",
"param_persistence_threshold": "auto",
"max_live_parameters": 0,
"max_reuse_distance": 0,
"gather_16bit_weights_on_model_save": true
},
"bf16": {
"enabled": "auto"
},
"fp16": {
"enabled": "auto",
"auto_cast": false,
"loss_scale": 0,
"initial_scale_power": 32,
"loss_scale_window": 1000,
"hysteresis": 2,
"min_loss_scale": 1
},
"gradient_accumulation_steps": "auto",
"gradient_clipping": "auto",
"train_batch_size": "auto",
"train_micro_batch_size_per_gpu": "auto",
"wall_clock_breakdown": false
}