Instructions to use stillerman/jason-expert-uspto-0.5k-same-ds with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use stillerman/jason-expert-uspto-0.5k-same-ds with Transformers:

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

pipe = pipeline("text-generation", model="stillerman/jason-expert-uspto-0.5k-same-ds")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("stillerman/jason-expert-uspto-0.5k-same-ds")
model = AutoModelForCausalLM.from_pretrained("stillerman/jason-expert-uspto-0.5k-same-ds")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use stillerman/jason-expert-uspto-0.5k-same-ds with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/stillerman/jason-expert-uspto-0.5k-same-ds

SGLang

How to use stillerman/jason-expert-uspto-0.5k-same-ds 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 "stillerman/jason-expert-uspto-0.5k-same-ds" \
    --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": "stillerman/jason-expert-uspto-0.5k-same-ds",
		"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 "stillerman/jason-expert-uspto-0.5k-same-ds" \
        --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": "stillerman/jason-expert-uspto-0.5k-same-ds",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use stillerman/jason-expert-uspto-0.5k-same-ds with Docker Model Runner:
```
docker model run hf.co/stillerman/jason-expert-uspto-0.5k-same-ds
```

layer_9,10,11,12,13

This model is a fine-tuned version of EleutherAI/pythia-1b-deduped on the Multi-Domain-Expert-Layers/uspto dataset. It achieves the following results on the evaluation set:

Loss: 2.2158
Accuracy: 0.5264

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: 0.0001
train_batch_size: 1
eval_batch_size: 8
seed: 42
distributed_type: multi-GPU
num_devices: 8
gradient_accumulation_steps: 8
total_train_batch_size: 64
total_eval_batch_size: 64
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
training_steps: 500

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
2.2702	0.01	200	2.2378	0.5232
2.2383	0.01	400	2.2208	0.5254

Framework versions

Transformers 4.28.1
Pytorch 2.0.1+rocm5.4.2
Datasets 2.11.0
Tokenizers 0.13.3

Wandb Report

https://wandb.ai/ontocord/jason-test-pythia-1b-deduped-layer-test-uspto/runs/c7pwmypg

Downloads last month: 3

Evaluation results

Accuracy on Multi-Domain-Expert-Layers/uspto
self-reported

0.526