Instructions to use lomahony/eleuther-pythia2.8b-hh-dpo with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use lomahony/eleuther-pythia2.8b-hh-dpo with Transformers:

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

pipe = pipeline("text-generation", model="lomahony/eleuther-pythia2.8b-hh-dpo")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("lomahony/eleuther-pythia2.8b-hh-dpo")
model = AutoModelForCausalLM.from_pretrained("lomahony/eleuther-pythia2.8b-hh-dpo")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use lomahony/eleuther-pythia2.8b-hh-dpo with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "lomahony/eleuther-pythia2.8b-hh-dpo"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "lomahony/eleuther-pythia2.8b-hh-dpo",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/lomahony/eleuther-pythia2.8b-hh-dpo

SGLang

How to use lomahony/eleuther-pythia2.8b-hh-dpo 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 "lomahony/eleuther-pythia2.8b-hh-dpo" \
    --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": "lomahony/eleuther-pythia2.8b-hh-dpo",
		"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 "lomahony/eleuther-pythia2.8b-hh-dpo" \
        --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": "lomahony/eleuther-pythia2.8b-hh-dpo",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use lomahony/eleuther-pythia2.8b-hh-dpo with Docker Model Runner:
```
docker model run hf.co/lomahony/eleuther-pythia2.8b-hh-dpo
```

Pythia-70m supervised finetuned with Anthropic-hh-rlhf dataset for 1 epoch (sft-model), before DPO (paper) with same dataset for 1 epoch.

wandb log

Benchmark evaluations included in repo done using lm-evaluation-harness.

See Pythia-70m for original model details (paper).

Downloads last month: 9

Dataset used to train lomahony/eleuther-pythia2.8b-hh-dpo

Collection including lomahony/eleuther-pythia2.8b-hh-dpo

Pythia-hh-all-sft-dpo

Collection

Pythia models supervised finetuned and DPO finetuned with all of Anthropic-hh-rlhf dataset for 1 epoch. • 13 items • Updated Mar 12, 2024

Papers for lomahony/eleuther-pythia2.8b-hh-dpo

Direct Preference Optimization: Your Language Model is Secretly a Reward Model

Paper • 2305.18290 • Published May 29, 2023 • 66

The Pile: An 800GB Dataset of Diverse Text for Language Modeling

Paper • 2101.00027 • Published Dec 31, 2020 • 10