Instructions to use Wojtekb30/Bielik-7B-Instruct-v0.1-Math-Code-adapter with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Wojtekb30/Bielik-7B-Instruct-v0.1-Math-Code-adapter with PEFT:

from peft import PeftModel
from transformers import AutoModelForCausalLM

base_model = AutoModelForCausalLM.from_pretrained("speakleash/Bielik-7B-Instruct-v0.1")
model = PeftModel.from_pretrained(base_model, "Wojtekb30/Bielik-7B-Instruct-v0.1-Math-Code-adapter")

Transformers

How to use Wojtekb30/Bielik-7B-Instruct-v0.1-Math-Code-adapter with Transformers:

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

pipe = pipeline("text-generation", model="Wojtekb30/Bielik-7B-Instruct-v0.1-Math-Code-adapter")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("Wojtekb30/Bielik-7B-Instruct-v0.1-Math-Code-adapter", dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use Wojtekb30/Bielik-7B-Instruct-v0.1-Math-Code-adapter with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Wojtekb30/Bielik-7B-Instruct-v0.1-Math-Code-adapter"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Wojtekb30/Bielik-7B-Instruct-v0.1-Math-Code-adapter",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Wojtekb30/Bielik-7B-Instruct-v0.1-Math-Code-adapter

SGLang

How to use Wojtekb30/Bielik-7B-Instruct-v0.1-Math-Code-adapter 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 "Wojtekb30/Bielik-7B-Instruct-v0.1-Math-Code-adapter" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Wojtekb30/Bielik-7B-Instruct-v0.1-Math-Code-adapter",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

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 "Wojtekb30/Bielik-7B-Instruct-v0.1-Math-Code-adapter" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Wojtekb30/Bielik-7B-Instruct-v0.1-Math-Code-adapter",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use Wojtekb30/Bielik-7B-Instruct-v0.1-Math-Code-adapter with Docker Model Runner:
```
docker model run hf.co/Wojtekb30/Bielik-7B-Instruct-v0.1-Math-Code-adapter
```

Bielik-7B-Instruct-v0.1-Math-Code-adapter / README.md

Wojtekb30

Update README.md

ecf4538 verified about 1 month ago

preview code

Raw

History Blame Contribute Delete

1.63 kB

metadata

license: apache-2.0
datasets:
  - ise-uiuc/Magicoder-Evol-Instruct-110K
  - ise-uiuc/Magicoder-OSS-Instruct-75K
  - meta-math/MetaMathQA
language:
  - en
  - pl
base_model:
  - speakleash/Bielik-7B-Instruct-v0.1
library_name: peft
pipeline_tag: text-generation
tags:
  - Math
  - Code
  - Adapter
  - base_model:adapter:speakleash/Bielik-7B-Instruct-v0.1
  - lora
  - sft
  - transformers
  - trl

Bielik-7B-Instruct-v0.1-Math-Code-adapter

An adapter for Bielik 7B Instruct v0.1 trained on English datasets:

https://huggingface.co/datasets/ise-uiuc/Magicoder-Evol-Instruct-110K

https://huggingface.co/datasets/ise-uiuc/Magicoder-OSS-Instruct-75K

https://huggingface.co/datasets/meta-math/MetaMathQA

Base model: https://huggingface.co/speakleash/Bielik-7B-Instruct-v0.1

It might be little overfitted.

It was part of my MoE experiment that I abandoned - I wanted to use this adapter to create one of the experts in a MoE LLM. I decided to publish it because it was already trained and I'd feel bad to waste it.

Adapter dla Bielik 7B Instruct v0.1 wytrenowany na anglojęzycznych datasetach:

https://huggingface.co/datasets/ise-uiuc/Magicoder-Evol-Instruct-110K

https://huggingface.co/datasets/ise-uiuc/Magicoder-OSS-Instruct-75K

https://huggingface.co/datasets/meta-math/MetaMathQA

Oryginalny model: https://huggingface.co/speakleash/Bielik-7B-Instruct-v0.1

Adapter mógł ulec overfittingowi.

Była to część mojego porzuconego eksperymentu z modelem MoE - chciałem wykorzystać ten adapter do stworzenia jednego z ekspertów w modelu MoE. Ponieważ adapter został już wytrenowany i szkoda byłoby go zmarnować, publikuję go.