configs/125M-moe.yml

# GPT-2 pretraining setup
{
   # See README for MoE config docs!
   "moe_type": "deepspeed",
   "moe_token_dropping": true,
   # Have 4 experts per layer (every 2 layers by default)
   "moe_num_experts": 4,
   # parallelism settings
   "enable_expert_tensor_parallelism": true,
   "pipe_parallel_size": 1, # not yet supported for MoE
   "model_parallel_size": 1,
   "moe_expert_parallel_size": 1,

   # model settings
   "num_layers": 12,
   "hidden_size": 768,
   "num_attention_heads": 12,
   "seq_length": 2048,
   "max_position_embeddings": 2048,
   "norm": "layernorm",
   "pos_emb": "rotary",
   "no_weight_tying": true,
   "gpt_j_residual": false,
   "output_layer_parallelism": "column",

   # these should provide some speedup but takes a while to build, set to true if desired
   "scaled_upper_triang_masked_softmax_fusion": false,
   "bias_gelu_fusion": false,
   "rope_fusion": false,

   # init methods
   "init_method": "small_init",
   "output_layer_init_method": "wang_init",


   # optimizer settings
   "optimizer": {
     "type": "Adam",
     "params": {
       "lr": 0.0006,
       "betas": [0.9, 0.95],
       "eps": 1.0e-8,
     }
   },
   "min_lr": 0.00006,

   # for all zero_optimization options, see https://www.deepspeed.ai/docs/config-json/#zero-optimizations-for-fp16-training
   "zero_optimization": {
    "stage": 1,
    "allgather_partitions": True,
    "allgather_bucket_size": 500000000,
    "overlap_comm": True,
    "reduce_scatter": True,
    "reduce_bucket_size": 500000000,
    "contiguous_gradients": True,
  },

   # batch / data settings
   "train_micro_batch_size_per_gpu": 4,
   "data_impl": "mmap",

   # activation checkpointing
   "checkpoint_activations": true,
   "checkpoint_num_layers": 1,
   "partition_activations": true,
   "synchronize_each_layer": true,

   # regularization
   "gradient_clipping": 1.0,
   "weight_decay": 0.1,
   "hidden_dropout": 0.0,
   "attention_dropout": 0.0,

   # precision settings
   "fp16": {
     "enabled": true,
     "loss_scale": 0,
     "loss_scale_window": 1000,
     "hysteresis": 2,
     "min_loss_scale": 1
   },

   # misc. training settings
   "train_iters": 320000,
   "lr_decay_iters": 320000,
   "distributed_backend": "nccl",
   "lr_decay_style": "cosine",
   "warmup": 0.01,
   "checkpoint_factor": 10000,
   "eval_interval": 1000,
   "eval_iters": 10,

   # logging
   "log_interval": 10,
   "steps_per_print": 10,
   "keep_last_n_checkpoints": 4,
   "wall_clock_breakdown": true,

  #  networking
  "hostfile": "/mock_path"
}