Instructions to use BramVanroy/llama2-13b-ft-mc4_nl_cleaned_tiny with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use BramVanroy/llama2-13b-ft-mc4_nl_cleaned_tiny with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="BramVanroy/llama2-13b-ft-mc4_nl_cleaned_tiny")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("BramVanroy/llama2-13b-ft-mc4_nl_cleaned_tiny")
model = AutoModelForCausalLM.from_pretrained("BramVanroy/llama2-13b-ft-mc4_nl_cleaned_tiny")

Inference
Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use BramVanroy/llama2-13b-ft-mc4_nl_cleaned_tiny with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "BramVanroy/llama2-13b-ft-mc4_nl_cleaned_tiny"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "BramVanroy/llama2-13b-ft-mc4_nl_cleaned_tiny",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/BramVanroy/llama2-13b-ft-mc4_nl_cleaned_tiny

SGLang

How to use BramVanroy/llama2-13b-ft-mc4_nl_cleaned_tiny with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "BramVanroy/llama2-13b-ft-mc4_nl_cleaned_tiny" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "BramVanroy/llama2-13b-ft-mc4_nl_cleaned_tiny",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "BramVanroy/llama2-13b-ft-mc4_nl_cleaned_tiny" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "BramVanroy/llama2-13b-ft-mc4_nl_cleaned_tiny",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use BramVanroy/llama2-13b-ft-mc4_nl_cleaned_tiny with Docker Model Runner:
```
docker model run hf.co/BramVanroy/llama2-13b-ft-mc4_nl_cleaned_tiny
```

llama2-13b-ft-mc4_nl_cleaned_tiny

This model is a fine-tuned version of meta-llama/Llama-2-13b-hf on the yhavinga/mc4_nl_cleaned dataset (tiny partition) on a context of 4096 tokens. See the original meta-llama/Llama-2-13b-hf for more information, intended use, and biases.

If you use this model or refer to it, please use the following citation:

Vanroy, B. (2023). Language Resources for Dutch Large Language Modelling. https://arxiv.org/abs/2312.12852

@article{vanroy2023language,
  title={Language Resources for {Dutch} Large Language Modelling},
  author={Vanroy, Bram},
  journal={arXiv preprint arXiv:2312.12852},
  year={2023}
}

Intended uses & limitations

While Llama 2 already contains some proficiency in Dutch, this finetune is intended to improve the fluency of Dutch (not increase its knowledge). It is therefore intended as a generative model for Dutch language. The biases, shortcomings and intended uses are otherwise the same as those of the original model. The model can be used for generative tasks or finetuned further on other tasks such as summarization, adaptation, instruction or chat finetuning.

Training and evaluation data

Trained on the yhavinga/mc4_nl_cleaned dataset (tiny partition) for one epoch. The canonical validation split was not used but instead 5% of train was used as validation.

Training procedure

Trained with LoRA targetting ["q_proj", "v_proj"] in 4 bit and merged before upload. Trained with Flash Attention as borrowed from here.

The adapters are in the adapters branch.

Initial training investigation on the Tier-1 HPC of Vlaams Supercomputer Centrum (VSC) and training on our own server of 4x 3090s.

Training hyperparameters

The following hyperparameters were used during training in the HPC investigation:

learning_rate: 0.0003
train_batch_size: 12
eval_batch_size: 12
seed: 42
distributed_type: multi-GPU
num_devices: 16
gradient_accumulation_steps: 6
total_train_batch_size: 1152
total_eval_batch_size: 192
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: cosine
lr_scheduler_warmup_ratio: 0.03
num_epochs: 1

Training results

Training Loss	Epoch	Step	Validation Loss
1.8784	0.09	90	1.8820
1.8344	0.19	180	1.8542
1.8351	0.28	270	1.8355
1.8206	0.37	360	1.8212
1.8021	0.47	450	1.8088
1.8102	0.56	540	1.7982
1.7991	0.65	630	1.7890
1.7788	0.74	720	1.7811
1.7915	0.84	810	1.7742
1.7715	0.93	900	1.7676

Framework versions

Transformers 4.31.0.dev0
Pytorch 2.0.1+cu117
Datasets 2.13.1
Tokenizers 0.13.3

Open LLM Leaderboard Evaluation Results

Detailed results can be found here

Metric	Value
Avg.	46.81
ARC (25-shot)	59.3
HellaSwag (10-shot)	82.04
MMLU (5-shot)	54.67
TruthfulQA (0-shot)	38.03
Winogrande (5-shot)	77.27
GSM8K (5-shot)	10.31
DROP (3-shot)	6.08