playground/deepspeed_config_s3.json

{
    "zero_optimization": {
        "stage": 3,
        "offload_optimizer": {
            "device": "cpu",
            "pin_memory": true
        },
        "offload_param": {
            "device": "cpu",
            "pin_memory": true
        },
        "overlap_comm": true,
        "contiguous_gradients": true,
        "sub_group_size": "auto",
        "reduce_bucket_size": "auto",
        "stage3_prefetch_bucket_size": "auto",
        "stage3_param_persistence_threshold": "auto",
        "stage3_max_live_parameters": "auto",
        "stage3_max_reuse_distance": "auto",
        "stage3_gather_16bit_weights_on_model_save": true
    },
    "optimizer": {
        "type": "AdamW",
        "params": {
            "lr": "auto",
            "betas": "auto",
            "eps": "auto",
            "weight_decay": "auto"
        }
    },
    "train_batch_size": "auto",
    "micro_batch_size_per_gpu": "auto",
    "gradient_accumulation_steps": "auto"
}