Examples

HANDS-ON USECASE EXAMPLES

Use Case	Description
Hugging Face Trainer	Train LLaMA 3-8B ~20% faster with over 40% memory reduction on Alpaca dataset using 4 A100s with FSDP
Lightning Trainer	Increase 15% throughput and reduce memory usage by 40% with LLaMA3-8B on MMLU dataset using 8 A100s with DeepSpeed ZeRO3
Medusa Multi-head LLM (Retraining Phase)	Reduce memory usage by 80% with 5 LM heads and improve throughput by 40% using 8 A100s with FSDP
Vision-Language Model SFT	Finetune Qwen2-VL on image-text data using 4 A100s with FSDP
Liger ORPO Trainer	Align Llama 3.2 using Liger ORPO Trainer with FSDP with 50% memory reduction

HuggingFace Trainer¶

How to Run¶

Locally on a GPU machine¶

You can run the example locally on a GPU machine. The default hyperparameters and configurations work on single node with 4xA100 80GB GPUs and FSDP.

Example

pip install -r requirements.txt
sh run_{MODEL}.sh

If you do not have access to a GPU machine, you can run the example on Modal. Modal is a serverless platform that allows you to run your code on a remote GPU machine. You can sign up for a free account at Modal.

Example

pip install modal
modal setup  # authenticate with Modal
modal run launch_on_modal.py --script "run_qwen2_vl.sh"

Notes

This example uses an optional use_liger flag. If true, it does a 1 line monkey patch to apply liger kernel.
The example uses Llama3 model that requires community license agreement and HuggingFace Hub login. If you want to use Llama3 in this example, please make sure you have done the following:
- Agree on the community license agreement .
- Run huggingface-cli login and enter your HuggingFace token.
The default hyperparameters and configurations work on single node with 4xA100 80GB GPUs. For running on device with less GPU RAM, please consider reducing the per-GPU batch size and/or enable CPUOffload in FSDP.

Benchmark Result¶

Llama¶

Info

Benchmark conditions: Model= LLaMA 3-8B,Datset= Alpaca, Max seq len = 512, Data Type = bf16, Optimizer = AdamW, Gradient Checkpointing = True, Distributed Strategy = FSDP1 on 4 A100s.

Throughput improves by around 20%, while GPU memory usage drops by 40%. This allows you to train the model on smaller GPUs, use larger batch sizes, or handle longer sequence lengths without incurring additional costs.

Qwen¶

Info

Benchmark conditions: Model= Qwen2-7B, Dataset= Alpaca, Max seq len = 512, Data Type = bf16, Optimizer = AdamW, Gradient Checkpointing = True, Distributed Strategy = FSDP1 on 4 A100s.

Throughput improves by around 10%, while GPU memory usage drops by 50%.

Gemma 7B¶

Info

Benchmark conditions: Model= Gemma-7B, Dataset= Alpaca, Max seq len = 512, Data Type = bf16, Optimizer = AdamW, Gradient Checkpointing = True, Distributed Strategy = FSDP1 on 4 A100s.

Throughput improves by around 24%, while GPU memory usage drops by 33%.

Lightning Trainer¶

How to Run¶

Locally on a GPU machine¶

You can run the example locally on a GPU machine.

Example

pip install -r requirements.txt

# For single L40 48GB GPU
python training.py --model Qwen/Qwen2-0.5B-Instruct --num_gpu 1 --max_length 1024

# For 8XA100 40GB
python training.py --model meta-llama/Meta-Llama-3-8B --strategy deepspeed

Notes

The example uses Llama3 model that requires community license agreement and HuggingFace Hub login. If you want to use Llama3 in this example, please make sure you have done the following:
- Agree on the community license agreement
- Run huggingface-cli login and enter your HuggingFace token.
The default hyperparameters and configurations for gemma works on single L40 48GB GPU and config for llama work on single node with 8xA100 40GB GPUs. For running on device with less GPU RAM, please consider reducing the per-GPU batch size and/or enable CPUOffload in FSDP.

Medusa¶

Medusa is a simple framework that democratizes the acceleration techniques for LLM generation with multiple decoding heads. To know more, you can check out the repo and the paper .

The Liger fused CE kernel is highly effective in this scenario, eliminating the need to materialize logits for each head, which usually consumes a large volume of memory due to the extensive vocabulary size (e.g., for LLaMA-3, the vocabulary size is 128k).

The introduction of multiple heads can easily lead to OOM (Out of Memory) issues. However, thanks to the efficient Liger fused CE, which calculates the gradient in place and doesn't materialize the logits, we have observed very effective results. This efficiency opens up more opportunities for multi-token prediction research and development.

How to Run¶

Example

git clone git@github.com:linkedin/Liger-Kernel.git
cd {PATH_TO_Liger-Kernel}/Liger-Kernel/
pip install -e .
cd {PATH_TO_Liger-Kernel}/Liger-Kernel/examples/medusa
pip install -r requirements.txt
sh scripts/llama3_8b_medusa.sh

Notes

This example uses an optional use_liger flag. If true, it does a monkey patch to apply liger kernel with medusa heads.
The example uses Llama3 model that requires community license agreement and HuggingFace Hub login. If you want to use Llama3 in this example, please make sure you have done the followings:
- Agree on the community license agreement https://huggingface.co/meta-llama/Meta-Llama-3-8B
- Run huggingface-cli login and enter your HuggingFace token
The default hyperparameters and configurations work on single node with 8xA100 GPUs. For running on device with less GPU RAM, please consider reducing the per-GPU batch size and/or enable CPUOffload in FSDP.
We are using a smaller sample of shared GPT data primarily to benchmark performance. The example requires hyperparameter tuning and dataset selection to work effectively, also ensuring the dataset has the same distribution as the LLaMA pretraining data. Welcome contribution to enhance the example code.

Benchmark Result¶

Info

Benchmark conditions: LLaMA 3-8B, Batch Size = 6, Data Type = bf16, Optimizer = AdamW, Gradient Checkpointing = True, Distributed Strategy = FSDP1 on 8 A100s.

Stage 1¶

Stage 1 refers to Medusa-1 where the backbone model is frozen and only weights of LLM heads are updated.

Warning

# Modify this flag in llama3_8b_medusa.sh to True enables stage1 
--medusa_only_heads True

num_head = 3¶

num_head = 5¶

Stage 2¶

Warning

# Modify this flag to False in llama3_8b_medusa.sh enables stage2
--medusa_only_heads False

Stage 2 refers to Medusa-2 where all the model weights are updated including the backbone model and llm heads.

num_head = 3¶

num_head = 5¶

Vision-Language Model SFT¶

How to Run¶

Locally on a GPU Machine¶

You can run the example locally on a GPU machine. The default hyperparameters and configurations work on single node with 4xA100 80GB GPUs.

Example

#!/bin/bash

torchrun --nnodes=1 --nproc-per-node=4 training_multimodal.py \
    --model_name "Qwen/Qwen2-VL-7B-Instruct" \
    --bf16 \
    --num_train_epochs 1 \
    --per_device_train_batch_size 8 \
    --per_device_eval_batch_size 8 \
    --eval_strategy "no" \
    --save_strategy "no" \
    --learning_rate 6e-6 \
    --weight_decay 0.05 \
    --warmup_ratio 0.1 \
    --lr_scheduler_type "cosine" \
    --logging_steps 1 \
    --include_num_input_tokens_seen \
    --report_to none \
    --fsdp "full_shard auto_wrap" \
    --fsdp_config config/fsdp_config.json \
    --seed 42 \
    --use_liger True \
    --output_dir multimodal_finetuning

ORPO Trainer¶

How to Run¶

Locally on a GPU Machine¶

You can run the example locally on a GPU machine and FSDP.

Example

import torch
from datasets import load_dataset
from transformers import AutoModelForCausalLM, AutoTokenizer
from trl import ORPOConfig  # noqa: F401

from liger_kernel.transformers.trainer import LigerORPOTrainer  # noqa: F401

model = AutoModelForCausalLM.from_pretrained(
    "meta-llama/Llama-3.2-1B-Instruct",
    torch_dtype=torch.bfloat16,
)

tokenizer = AutoTokenizer.from_pretrained(
    "meta-llama/Llama-3.2-1B-Instruct",
    max_length=512,
    padding="max_length",
)
tokenizer.pad_token = tokenizer.eos_token

train_dataset = load_dataset("trl-lib/tldr-preference", split="train")

training_args = ORPOConfig(
    output_dir="Llama3.2_1B_Instruct",
    beta=0.1,
    max_length=128,
    per_device_train_batch_size=32,
    max_steps=100,
    save_strategy="no",
)

trainer = LigerORPOTrainer(
    model=model, args=training_args, tokenizer=tokenizer, train_dataset=train_dataset
)

trainer.train()

Examples

HuggingFace Trainer¶

How to Run¶

Locally on a GPU machine¶

Remotely on Modal¶

Benchmark Result¶

Llama¶

Qwen¶

Gemma 7B¶

Lightning Trainer¶

How to Run¶

Locally on a GPU machine¶

Medusa¶

How to Run¶

Benchmark Result¶

Stage 1¶

num_head = 3¶

num_head = 5¶

Stage 2¶

num_head = 3¶

num_head = 5¶

Vision-Language Model SFT¶

How to Run¶

Locally on a GPU Machine¶

ORPO Trainer¶

How to Run¶

Locally on a GPU Machine¶