Instructions to use CompassioninMachineLearning/llama-3.2-1b-paw-control-lora with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use CompassioninMachineLearning/llama-3.2-1b-paw-control-lora with PEFT:

from peft import PeftModel
from transformers import AutoModelForCausalLM

base_model = AutoModelForCausalLM.from_pretrained("meta-llama/Llama-3.2-1B")
model = PeftModel.from_pretrained(base_model, "CompassioninMachineLearning/llama-3.2-1b-paw-control-lora")

Transformers

How to use CompassioninMachineLearning/llama-3.2-1b-paw-control-lora with Transformers:

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

pipe = pipeline("text-generation", model="CompassioninMachineLearning/llama-3.2-1b-paw-control-lora")

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("CompassioninMachineLearning/llama-3.2-1b-paw-control-lora", dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use CompassioninMachineLearning/llama-3.2-1b-paw-control-lora with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "CompassioninMachineLearning/llama-3.2-1b-paw-control-lora"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "CompassioninMachineLearning/llama-3.2-1b-paw-control-lora",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/CompassioninMachineLearning/llama-3.2-1b-paw-control-lora

SGLang

How to use CompassioninMachineLearning/llama-3.2-1b-paw-control-lora 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 "CompassioninMachineLearning/llama-3.2-1b-paw-control-lora" \
    --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": "CompassioninMachineLearning/llama-3.2-1b-paw-control-lora",
		"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 "CompassioninMachineLearning/llama-3.2-1b-paw-control-lora" \
        --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": "CompassioninMachineLearning/llama-3.2-1b-paw-control-lora",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use CompassioninMachineLearning/llama-3.2-1b-paw-control-lora with Docker Model Runner:
```
docker model run hf.co/CompassioninMachineLearning/llama-3.2-1b-paw-control-lora
```

llama-3.2-1b-paw-control-lora / tokenizer_config.json

sparrow8i8

Random Wikipedia content LoRA adapter (Llama-3.2-1B, r=32, 36 random docs, 3 epochs)

2202837 verified about 1 month ago

raw

history blame contribute delete

335 Bytes

	{
	"backend": "tokenizers",
	"bos_token": "<\|begin_of_text\|>",
	"clean_up_tokenization_spaces": true,
	"eos_token": "<\|end_of_text\|>",
	"is_local": false,
	"model_input_names": [
	"input_ids",
	"attention_mask"
	],
	"model_max_length": 131072,
	"pad_token": "<\|end_of_text\|>",
	"tokenizer_class": "TokenizersBackend"
	}