Instructions to use jbenbudd/ADPrLlama with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use jbenbudd/ADPrLlama with Transformers:

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

pipe = pipeline("text-generation", model="jbenbudd/ADPrLlama")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("jbenbudd/ADPrLlama")
model = AutoModelForCausalLM.from_pretrained("jbenbudd/ADPrLlama", device_map="auto")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=256)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

PEFT
How to use jbenbudd/ADPrLlama with PEFT:
```
Task type is invalid.
```
Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use jbenbudd/ADPrLlama with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "jbenbudd/ADPrLlama"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "jbenbudd/ADPrLlama",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/jbenbudd/ADPrLlama

SGLang

How to use jbenbudd/ADPrLlama 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 "jbenbudd/ADPrLlama" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "jbenbudd/ADPrLlama",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

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 "jbenbudd/ADPrLlama" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "jbenbudd/ADPrLlama",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use jbenbudd/ADPrLlama with Docker Model Runner:
```
docker model run hf.co/jbenbudd/ADPrLlama
```

ADPrLlama / training_args.yaml

Ben Buddendorff

Initial commit of the LoRA/adapter model

7ba0b05 over 1 year ago

Raw History Blame Contribute Delete

871 Bytes

	bf16: true
	cutoff_len: 2048
	dataset: adpr_train
	dataset_dir: data
	ddp_timeout: 180000000
	do_train: true
	eval_steps: 100
	eval_strategy: steps
	finetuning_type: lora
	flash_attn: auto
	gradient_accumulation_steps: 8
	include_num_input_tokens_seen: true
	learning_rate: 5.0e-05
	logging_steps: 5
	lora_alpha: 128
	lora_dropout: 0.01
	lora_rank: 64
	lora_target: q_proj,v_proj,k_proj,o_proj,gate_proj,down_proj,up_proj
	lr_scheduler_type: cosine
	max_grad_norm: 1.0
	max_samples: 100000
	model_name_or_path: GreatCaptainNemo/ProLLaMA
	num_train_epochs: 3.0
	optim: adamw_torch
	output_dir: saves/Custom/lora/train_2025-04-05-23-57-03
	packing: false
	per_device_eval_batch_size: 16
	per_device_train_batch_size: 16
	plot_loss: true
	preprocessing_num_workers: 16
	report_to: none
	resize_vocab: true
	save_steps: 100
	stage: sft
	template: alpaca
	trust_remote_code: true
	val_size: 0.1
	warmup_steps: 20