Instructions to use OpenPipe/sample-lora-pii-redaction with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use OpenPipe/sample-lora-pii-redaction with Transformers:

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

pipe = pipeline("text-generation", model="OpenPipe/sample-lora-pii-redaction")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("OpenPipe/sample-lora-pii-redaction")
model = AutoModelForCausalLM.from_pretrained("OpenPipe/sample-lora-pii-redaction")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use OpenPipe/sample-lora-pii-redaction with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/OpenPipe/sample-lora-pii-redaction

SGLang

How to use OpenPipe/sample-lora-pii-redaction 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 "OpenPipe/sample-lora-pii-redaction" \
    --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": "OpenPipe/sample-lora-pii-redaction",
		"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 "OpenPipe/sample-lora-pii-redaction" \
        --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": "OpenPipe/sample-lora-pii-redaction",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use OpenPipe/sample-lora-pii-redaction with Docker Model Runner:
```
docker model run hf.co/OpenPipe/sample-lora-pii-redaction
```

sample-lora-pii-redaction / README.md

corbt

Initial commit

caab2a0 over 2 years ago

preview code

raw

history blame contribute delete

2.25 kB

	---
	license: apache-2.0
	base_model: OpenPipe/mistral-ft-optimized-1218
	tags:
	- generated_from_trainer
	model-index:
	- name: models/loras/OpenPipe/ft-development-0b0f52d6-bc53-4443-bbad-4a6103c95501-pii-7b-optimized
	results: []
	---

	<!-- This model card has been generated automatically according to the information the Trainer had access to. You
	should probably proofread and complete it, then remove this comment. -->

	[<img src="https://raw.githubusercontent.com/OpenAccess-AI-Collective/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/OpenAccess-AI-Collective/axolotl)
	# models/loras/OpenPipe/ft-development-0b0f52d6-bc53-4443-bbad-4a6103c95501-pii-7b-optimized

	This model is a fine-tuned version of [OpenPipe/mistral-ft-optimized-1218](https://huggingface.co/OpenPipe/mistral-ft-optimized-1218) on the None dataset.
	It achieves the following results on the evaluation set:
	- Loss: 0.0174

	## Model description

	More information needed

	## Intended uses & limitations

	More information needed

	## Training and evaluation data

	More information needed

	## Training procedure

	### Training hyperparameters

	The following hyperparameters were used during training:
	- learning_rate: 0.0002
	- train_batch_size: 2
	- eval_batch_size: 2
	- seed: 42
	- gradient_accumulation_steps: 4
	- total_train_batch_size: 8
	- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
	- lr_scheduler_type: cosine
	- lr_scheduler_warmup_steps: 10
	- num_epochs: 2

	### Training results

	\| Training Loss \| Epoch \| Step \| Validation Loss \|
	\|:-------------:\|:-----:\|:----:\|:---------------:\|
	\| 0.42 \| 0.02 \| 1 \| 0.3989 \|
	\| 0.0258 \| 0.21 \| 13 \| 0.0304 \|
	\| 0.025 \| 0.43 \| 26 \| 0.0220 \|
	\| 0.0146 \| 0.64 \| 39 \| 0.0204 \|
	\| 0.0208 \| 0.85 \| 52 \| 0.0196 \|
	\| 0.0136 \| 1.07 \| 65 \| 0.0187 \|
	\| 0.0148 \| 1.28 \| 78 \| 0.0181 \|
	\| 0.0178 \| 1.49 \| 91 \| 0.0180 \|
	\| 0.0204 \| 1.7 \| 104 \| 0.0175 \|
	\| 0.0128 \| 1.92 \| 117 \| 0.0174 \|


	### Framework versions

	- Transformers 4.34.1
	- Pytorch 2.0.1+cu117
	- Datasets 2.14.6
	- Tokenizers 0.14.1