Instructions to use pvduy/ppo_pythia6B_sample with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use pvduy/ppo_pythia6B_sample with Transformers:

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

pipe = pipeline("text-generation", model="pvduy/ppo_pythia6B_sample")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("pvduy/ppo_pythia6B_sample")
model = AutoModelForCausalLM.from_pretrained("pvduy/ppo_pythia6B_sample")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use pvduy/ppo_pythia6B_sample with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "pvduy/ppo_pythia6B_sample"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "pvduy/ppo_pythia6B_sample",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/pvduy/ppo_pythia6B_sample

SGLang

How to use pvduy/ppo_pythia6B_sample 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 "pvduy/ppo_pythia6B_sample" \
    --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": "pvduy/ppo_pythia6B_sample",
		"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 "pvduy/ppo_pythia6B_sample" \
        --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": "pvduy/ppo_pythia6B_sample",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use pvduy/ppo_pythia6B_sample with Docker Model Runner:
```
docker model run hf.co/pvduy/ppo_pythia6B_sample
```

YAML Metadata Warning:empty or missing yaml metadata in repo card

Check out the documentation for more information.

Inference code:

from datasets import load_dataset

from transformers import AutoTokenizer, AutoModelForCausalLM

dataset = load_dataset("CarperAI/openai_summarize_tldr")

val_prompts = [sample["prompt"] for sample in dataset["valid"]]

kwargs = {
    "max_new_tokens": 50,
    "do_sample": True,
    "top_k": 0,
    "top_p": 1,
}

model = AutoModelForCausalLM.from_pretrained("pvduy/ppo_pythia6B_sample")
model.eval()
tokenizer = AutoTokenizer.from_pretrained("pvduy/ppo_pythia6B_sample")
tokenizer.pad_token_id = tokenizer.eos_token_id

count = 0

for prompt in val_prompts:
    output_tk = tokenizer(prompt, return_tensors="pt")
    outputs = model.generate(output_tk.input_ids, attention_mask=output_tk.attention_mask, **kwargs)
    print("Prompt:", prompt)
    print("Output:", tokenizer.decode(outputs[0], skip_special_tokens=True).split("TL;DR:")[1].strip())
    print("=================================")
    count += 1
    if count == 10:
        break

Downloads last month: 7