# Copyright (c) 2024, EleutherAI
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
#     http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.

# GPT_2 pretraining setup
{
   # parallelism settings ( you will want to change these based on your cluster setup, ideally scheduling pipeline stages
   # across the node boundaries )
   "pipe_parallel_size": 0,
   "model_parallel_size": 1,

   # model settings
   "num_layers": 2,
   "hidden_size": 192,
   "num_attention_heads": 6,
   "seq_length": 1024,
   "max_position_embeddings": 1024,
   "norm": "layernorm",
   "pos_emb": "rotary",
   "no_weight_tying": true,

   # 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,
   "layernorm_fusion": false,

   # optimizer settings
   "optimizer": {
     "type": "Adam",
     "params": {
       "lr": 0.0006,
       "betas": [0.9, 0.999],
       "eps": 1.0e-8,
     }
   },

   # for all zero_optimization options, see https://www.deepspeed.ai/docs/config-json/#zero-optimizations-for-fp16-training
   "zero_optimization": {
    "stage": 0,
    "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",
   "split": "949,50,1",

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

   # regularization
   "gradient_clipping": 1.0,
   "weight_decay": 0.0,
   "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": 100,
   "steps_per_print": 10,
   "keep_last_n_checkpoints": 4,
   "wall_clock_breakdown": true,

  # Suggested data paths when using GPT_NeoX locally
  "data_path": "data/enwik8/enwik8_text_document",

  # or for weighted datasets:
  # "train-data-paths": ["data/enwik8/enwik8_text_document", "data/enwik8/enwik8_text_document"],
  # "test-data-paths": ["data/enwik8/enwik8_text_document", "data/enwik8/enwik8_text_document"],
  # "valid-data-paths": ["data/enwik8/enwik8_text_document", "data/enwik8/enwik8_text_document"],
  # "train-data-weights": [1., 2.],
  # "test-data-weights": [2., 1.],
  # "valid-data-weights": [0.5, 0.4],

  "vocab_file": "data/gpt2-vocab.json",
  "merge_file": "data/gpt2-merges.txt",
  "save": "test_checkpoint",
  "load": "test_checkpoint",
  "tensorboard_dir": "test_tensorboard",
  "log_dir": "test_logs",

}