Instructions to use instruction-pretrain/instruction-synthesizer with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use instruction-pretrain/instruction-synthesizer with Transformers:

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

pipe = pipeline("text-generation", model="instruction-pretrain/instruction-synthesizer")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("instruction-pretrain/instruction-synthesizer")
model = AutoModelForCausalLM.from_pretrained("instruction-pretrain/instruction-synthesizer")

Inference
Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use instruction-pretrain/instruction-synthesizer with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/instruction-pretrain/instruction-synthesizer

SGLang

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

Docker Model Runner
How to use instruction-pretrain/instruction-synthesizer with Docker Model Runner:
```
docker model run hf.co/instruction-pretrain/instruction-synthesizer
```

Update README.md

by AdaptLLM - opened Aug 1, 2024

base: refs/heads/main

←

from: refs/pr/3

Discussion Files changed

-1

Files changed (1) hide show

README.md +1 -1

README.md CHANGED Viewed

@@ -186,7 +186,7 @@ Except for the pre-training data, *Instruction Pre-Training* keeps all other set
 Therefore, you can easily use any training framework, such as [OLMo](https://github.com/allenai/OLMo) (for pre-training from scratch) and [LLaMA-Factory](https://github.com/hiyouga/LLaMA-Factory) (for continual pre-training), to train on the templified instruction-augmented corpora.
 1. For general pre-training from scratch, we recommend setting M = 2 and mixing the instruction-augmented corpora with unchanged raw corpora.
-2. For domain-adaptive continual pre-training, we recommend setting M = 3 and mixing the instruction-augmented corpora with general instructions from [OpenOrca](https://huggingface.co/datasets/Open-Orca/OpenOrca) at a 1:1 ratio (counted by tokens). Each example from OpenOrca is formulated as "{question} {response}", with a blank space used to connect the question and response.
 Let's try our method in continual pre-training for a quick start---it works easily!

 Therefore, you can easily use any training framework, such as [OLMo](https://github.com/allenai/OLMo) (for pre-training from scratch) and [LLaMA-Factory](https://github.com/hiyouga/LLaMA-Factory) (for continual pre-training), to train on the templified instruction-augmented corpora.
 1. For general pre-training from scratch, we recommend setting M = 2 and mixing the instruction-augmented corpora with unchanged raw corpora.
+2. For domain-adaptive continual pre-training, we recommend setting M = 3 and mixing the instruction-augmented corpora with general instructions from [OpenOrca](https://huggingface.co/datasets/Open-Orca/OpenOrca) at a 1:1 ratio (counted by tokens). Each example from OpenOrca is formulated as "{question} {response}", with a white-space used to connect the question and response.
 Let's try our method in continual pre-training for a quick start---it works easily!