Instructions to use HWERI/pythia-70m-deduped-cleansharegpt with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use HWERI/pythia-70m-deduped-cleansharegpt with Transformers:

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

pipe = pipeline("text-generation", model="HWERI/pythia-70m-deduped-cleansharegpt")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("HWERI/pythia-70m-deduped-cleansharegpt")
model = AutoModelForCausalLM.from_pretrained("HWERI/pythia-70m-deduped-cleansharegpt")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use HWERI/pythia-70m-deduped-cleansharegpt with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "HWERI/pythia-70m-deduped-cleansharegpt"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "HWERI/pythia-70m-deduped-cleansharegpt",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/HWERI/pythia-70m-deduped-cleansharegpt

SGLang

How to use HWERI/pythia-70m-deduped-cleansharegpt 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 "HWERI/pythia-70m-deduped-cleansharegpt" \
    --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": "HWERI/pythia-70m-deduped-cleansharegpt",
		"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 "HWERI/pythia-70m-deduped-cleansharegpt" \
        --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": "HWERI/pythia-70m-deduped-cleansharegpt",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use HWERI/pythia-70m-deduped-cleansharegpt with Docker Model Runner:
```
docker model run hf.co/HWERI/pythia-70m-deduped-cleansharegpt
```

Model Card

Pythia-70m-deduped finetuned on a cleaned version of ShareGPT data.

Open LLM Leaderboard Evaluation Results

Detailed results can be found here

Metric	Value
Avg.	25.34
ARC (25-shot)	25.68
HellaSwag (10-shot)	25.4
MMLU (5-shot)	23.12
TruthfulQA (0-shot)	51.15
Winogrande (5-shot)	52.01
GSM8K (5-shot)	0.0
DROP (3-shot)	0.0

Downloads last month: 912

Safetensors

Model size

70.4M params

Tensor type

F32

HWERI
/

pythia-70m-deduped-cleansharegpt

Model Card

Open LLM Leaderboard Evaluation Results

Dataset used to train HWERI/pythia-70m-deduped-cleansharegpt