scripts/pretrain-model.yaml

# https://huggingface.co/meta-llama/Llama-3.2-1B-Instruct/blob/main/config.json

# The name of the model to pretrain. Choose from names in ``litgpt.config``. Mutually exclusive with
# ``model_config``. (type: Optional[str], default: null)
model_name: "Llama-3.1-8B"

# A ``litgpt.Config`` object to define the model architecture. Mutually exclusive with
# ``model_config``. (type: Optional[Config], default: null)
model_config:
  padded_vocab_size: 262144
  vocab_size: 262144
  block_size: 8192
  n_layer: 32
  n_head: 32
  head_size: 64
  n_embd: 256
  n_query_groups: 8
  rotary_percentage: 1.0
  parallel_residual: false
  shared_attention_norm: false
  bias: false
  norm_class_name: "RMSNorm"
  norm_eps: 1e-05
  mlp_class_name: "LLaMAMLP"
  intermediate_size: 1024
  rope_base: 500000
  rope_adjustments:
    factor: 32.0
    low_freq_factor: 1.0
    high_freq_factor: 4.0
    original_max_seq_len: 8192

# Directory in which to save checkpoints and logs. If running in a Lightning Studio Job, look for it in
# /teamspace/jobs/<job-name>/share. (type: <class 'Path'>, default: out/pretrain)
out_dir: "../out/pretrain/"

# The precision to use for pretraining. Possible choices: "bf16-true", "bf16-mixed", "32-true". (type: Optional[str], default: null)
# precision: bf16-mixed
precision: bf16-true

# Optional path to a checkpoint directory to initialize the model from.
# Useful for continued pretraining. Mutually exclusive with ``resume``. (type: Optional[Path], default: null)
initial_checkpoint_dir:

# Path to a checkpoint directory to resume from in case training was interrupted, or ``True`` to resume
# from the latest checkpoint in ``out_dir``. An error will be raised if no checkpoint is found. Passing
# ``'auto'`` will resume from the latest checkpoint but not error if no checkpoint exists.
# (type: Union[bool, Literal["auto"], Path], default: False)
# resume: false
resume: "auto"

# Data-related arguments. If not provided, the default is ``litgpt.data.TinyLlama``.
data:
  class_path: LitData

  init_args:
    data_path: "../pretrain-data/"
    num_workers: 32

# Training-related arguments. See ``litgpt.args.TrainArgs`` for details
train:
  # Number of optimizer steps between saving checkpoints (type: Optional[int], default: 1000)
  save_interval: 500

  # Number of iterations between logging calls (type: int, default: 1)
  log_interval: 1

  # Number of samples between optimizer steps across data-parallel ranks (type: int, default: 512)
  global_batch_size: 512

  # Number of samples per data-parallel rank (type: int, default: 4)
  micro_batch_size: 11

  # Number of iterations with learning rate warmup active (type: int, default: 2000)
  lr_warmup_steps: 2000

  # Number of epochs to train on (type: Optional[int], default: null)
  epochs:

  # Total number of tokens to train on (type: Optional[int], default: 3000000000000)
  max_tokens: 9889496064 # 9657711 * 512 * 2
  # max_tokens: 4944748032 # 9657711 * 512 * 1

  # Limits the number of optimizer steps to run. (type: Optional[int], default: null)
  max_steps:

  # Limits the length of samples. Off by default (type: Optional[int], default: null)
  max_seq_length: 512

  # Whether to tie the embedding weights with the language modeling head weights. (type: Optional[bool], default: False)
  tie_embeddings: true

  #   (type: Optional[float], default: 1.0)
  max_norm: 1.0

  #   (type: float, default: 4e-05)
  min_lr: 4e-05

# Evaluation-related arguments. See ``litgpt.args.EvalArgs`` for details
eval:
  # Number of optimizer steps between evaluation calls (type: int, default: 1000)
  interval: 100

  # Number of tokens to generate (type: Optional[int], default: null)
  max_new_tokens:

  # Number of iterations (type: int, default: 100)
  max_iters: 100

  # Whether to evaluate on the validation set at the beginning of the training
  initial_validation: false

  # Whether to evaluate on the validation set at the end the training
  final_validation: true

# Optimizer-related arguments
optimizer:
  # class_path: torch.optim.AdamW
  class_path: grokadamw.GrokAdamW

  init_args:
    #   (type: float, default: 0.001)
    lr: 1e-3

    #   (type: float, default: 0.01)
    weight_decay: 1e-2

    #   (type: tuple, default: (0.9,0.999))
    betas:
      - 0.9
      - 0.999

# How many devices/GPUs to use. Uses all GPUs by default. (type: Union[int, str], default: auto)
devices: auto

# How many nodes to use. (type: int, default: 1)
num_nodes: 1

# Optional path to the tokenizer dir that was used for preprocessing the dataset. Only some data
# module require this. (type: Optional[Path], default: null)
tokenizer_dir: "../"

# The name of the logger to send metrics to. (type: Literal['wandb', 'tensorboard', 'csv'], default: tensorboard)
logger_name: "wandb"

# The random seed to use for reproducibility. (type: int, default: 42)
seed: 23