Important

NeMo 2.0 is an experimental feature and currently released in the dev container only: nvcr.io/nvidia/nemo:dev. Please refer to NeMo 2.0 overview for information on getting started.

Training with Predefined Configurations

NVIDIA provides configurations for two Baichuan2 model sizes: 7B, 13B.

Run Training

To run Baichuan2 training update conf/config.yaml:

defaults:
  - training: baichuan2/baichuan2_7b

stages:
  - training

Specify baichuan2 and the desired model size for training configuration, baichuan2/baichuan2_<model_size>.

Execute the launcher pipeline: python3 main.py.

Configuration

Default configurations for model size specific training can be found in the folder conf/training/baichuan2. The configuration is divided into four sections run, trainer, exp_manager, and model.

run:
  name: baichuan2_7b
  results_dir: ${base_results_dir}/${.name}
  time_limit: "0-04:00:00"
  dependency: "singleton"

Set the number of nodes and devices for training:

trainer:
  num_nodes: 16
  devices: 8
  max_steps: 300000 # consumed_samples = global_step * global_batch_size
  max_time: "05:23:30:00" # days:hours:minutes:seconds

Set configurations for creating a checkpoint:

exp_manger:
  create_checkpoint_callback: True
  checkpoint_callback_params:
    monitor: val_loss
    save_top_k: 10
    mode: min
    always_save_nemo: False # saves nemo file during validation, not implemented for model parallel
    save_nemo_train_end: False # not recommended when training large models on clusters with short time limits
    filename: 'megatron_baichuan2--{val_loss:.2f}-{step}-{consumed_amples}'
    model_parallel_size: ${multiply:${training.model.tensor_model_parallel_size}, ${training.model.pipeline_model_parallel_size}}

Set wandb configurations:

exp_manager:
  create_wandb_logger: True
  wandb_logger_kwargs:
    project: nemo_baichuan2
    name: ${training.run.name}

Set tensor parallel and pipeline parallel size:

model:
  tensor_model_parallel_size: 1
  pipeline_model_parallel_size: 1

Set data distribution configuration:

model:
  data:
    data_prefix:
    - .0333
    - ${data_dir}/my-baichuan2_00_text_document
    - .0333
    - ${data_dir}/my-baichuan2_00_text_document