Instructions to use lvwerra/starcoderbase-gsm8k with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use lvwerra/starcoderbase-gsm8k with Transformers:

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

pipe = pipeline("text-generation", model="lvwerra/starcoderbase-gsm8k")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("lvwerra/starcoderbase-gsm8k")
model = AutoModelForCausalLM.from_pretrained("lvwerra/starcoderbase-gsm8k")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use lvwerra/starcoderbase-gsm8k with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "lvwerra/starcoderbase-gsm8k"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "lvwerra/starcoderbase-gsm8k",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/lvwerra/starcoderbase-gsm8k

SGLang

How to use lvwerra/starcoderbase-gsm8k 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 "lvwerra/starcoderbase-gsm8k" \
    --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": "lvwerra/starcoderbase-gsm8k",
		"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 "lvwerra/starcoderbase-gsm8k" \
        --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": "lvwerra/starcoderbase-gsm8k",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use lvwerra/starcoderbase-gsm8k with Docker Model Runner:
```
docker model run hf.co/lvwerra/starcoderbase-gsm8k
```

lvwerra HF Staff

vwxyzjn commited on Aug 30, 2023

Commit

44fd019

1 Parent(s): c3a7a19

Create README.md (#2)

Browse files

- Create README.md (c6f16a6d20f04cd75bdaa264d4251e8b07184b18)

Co-authored-by: Shengyi Costa Huang <vwxyzjn@users.noreply.huggingface.co>

Files changed (1) hide show

README.md +32 -0

README.md ADDED Viewed

	@@ -0,0 +1,32 @@

+---
+license: bigscience-openrail-m
+datasets:
+- gsm8k
+language:
+- en
+tags:
+- trl
+- transformers
+- rlhf
+---
+# starcoderbase-triviaqa
+This model is baesed on https://huggingface.co/bigcode/starcoderbase and is fine-tuned on the GSM8K dataset using reinforcement learning via TRL's `TextEnvironment` (https://github.com/huggingface/trl/pull/424).
+### Out of Scope Use
+- Replacing human expertise
+## Bias, Risks, and Limitations
+- May generate answers that are incorrect or misleading.
+- May copy answers from the training data verbatim.
+- May generate language that is hateful or promotes discrimination ([example](https://huggingface.co/trl-lib/llama-7b-se-rl-peft/discussions/7#64376083369f6f907f5bfe4c)).
+- May generate language that is offensive to direct or indirect users or to people or groups mentioned.
+### Recommendations
+- Answers should be validated through the use of external sources.
+- Disparities between the data contributors and the direct and indirect users of the technology should inform developers in assessing what constitutes an appropriate use case.
+- Further research is needed to attribute model generations to sources in the training data, especially in cases where the model copies answers from the training data.