3B_qat_lora.yaml

# Config for multi-device QAT + LoRA finetuning in qat_lora_finetune_distributed.py
# using a Llama3.2 3B Instruct model
#
# This config assumes that you've run the following command before launching
# this run:
#   tune download meta-llama/Llama-3.2-3B-Instruct --output-dir /tmp/Llama-3.2-3B-Instruct --ignore-patterns "original/consolidated.00.pth"
#
# To launch on 2 devices, run the following command from root:
#   tune run --nproc_per_node 2 qat_lora_finetune_distributed --config llama3_2/3B_qat_lora
#
# You can add specific overrides through the command line. For example
# to override the checkpointer directory while launching training
# you can run:
#   tune run --nproc_per_node 2 qat_lora_finetune_distributed --config llama3_2/3B_qat_lora checkpointer.checkpoint_dir=<YOUR_CHECKPOINT_DIR>

output_dir: /tmp/torchtune/llama3_2_3B/qat_lora # /tmp may be deleted by your system. Change it to your preference.

# Tokenizer
tokenizer:
  _component_: torchtune.models.llama3.llama3_tokenizer
  path: /tmp/Llama-3.2-3B-Instruct/original/tokenizer.model
  max_seq_len: null

# Model Arguments
model:
  _component_: torchtune.models.llama3_2.lora_llama3_2_3b
  lora_attn_modules: ['q_proj', 'v_proj', 'output_proj']
  apply_lora_to_mlp: True
  lora_rank: 64  # higher increases accuracy and memory
  lora_alpha: 128  # usually alpha=2*rank
  lora_dropout: 0.0

checkpointer:
  _component_: torchtune.training.FullModelHFCheckpointer
  checkpoint_dir: /tmp/Llama-3.2-3B-Instruct/
  checkpoint_files: [
    model-00001-of-00002.safetensors,
    model-00002-of-00002.safetensors,
  ]
  recipe_checkpoint: null
  output_dir: ${output_dir}
  model_type: LLAMA3_2
resume_from_checkpoint: False
save_adapter_weights_only: False

# Dataset and Sampler
dataset:
  _component_: torchtune.datasets.alpaca_cleaned_dataset
  packed: False  # True increases speed
seed: null
shuffle: True
batch_size: 4

# Optimizer and Scheduler
optimizer:
  _component_: torch.optim.AdamW
  fused: True
  weight_decay: 0.01
  lr: 3e-4
lr_scheduler:
  _component_: torchtune.training.lr_schedulers.get_cosine_schedule_with_warmup
  num_warmup_steps: 100

loss:
  _component_: torchtune.modules.loss.CEWithChunkedOutputLoss

# Training
epochs: 1
max_steps_per_epoch: null
gradient_accumulation_steps: 8  # Use to increase effective batch size
clip_grad_norm: null
compile: False  # torch.compile the model + loss, True increases speed + decreases memory

# Logging
metric_logger:
  _component_: torchtune.training.metric_logging.DiskLogger
  log_dir: ${output_dir}/logs
log_every_n_steps: 1
log_peak_memory_stats: True

# Environment
device: cuda
dtype: bf16
enable_activation_checkpointing: False  # True reduces memory
enable_activation_offloading: False  # True reduces memory

# QAT arguments
quantizer:
  _component_: torchtune.training.quantization.Int8DynActInt4WeightQATQuantizer
  groupsize: 256

# Profiler (disabled)
profiler:
  _component_: torchtune.training.setup_torch_profiler
  enabled: False

  #Output directory of trace artifacts
  output_dir: ${output_dir}/profiling_outputs

  #`torch.profiler.ProfilerActivity` types to trace
  cpu: True
  cuda: True

  #trace options passed to `torch.profiler.profile`
  profile_memory: False
  with_stack: False
  record_shapes: True
  with_flops: False

  # `torch.profiler.schedule` options:
  # wait_steps -> wait, warmup_steps -> warmup, active_steps -> active, num_cycles -> repeat
  wait_steps: 5
  warmup_steps: 3
  active_steps: 2
  num_cycles: 1