Instructions to use OpenNLPLab/TransNormerLLM2-7B-300B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use OpenNLPLab/TransNormerLLM2-7B-300B with Transformers:

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

pipe = pipeline("text-generation", model="OpenNLPLab/TransNormerLLM2-7B-300B", trust_remote_code=True)

# Load model directly
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("OpenNLPLab/TransNormerLLM2-7B-300B", trust_remote_code=True, dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use OpenNLPLab/TransNormerLLM2-7B-300B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "OpenNLPLab/TransNormerLLM2-7B-300B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "OpenNLPLab/TransNormerLLM2-7B-300B",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/OpenNLPLab/TransNormerLLM2-7B-300B

SGLang

How to use OpenNLPLab/TransNormerLLM2-7B-300B 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 "OpenNLPLab/TransNormerLLM2-7B-300B" \
    --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": "OpenNLPLab/TransNormerLLM2-7B-300B",
		"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 "OpenNLPLab/TransNormerLLM2-7B-300B" \
        --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": "OpenNLPLab/TransNormerLLM2-7B-300B",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use OpenNLPLab/TransNormerLLM2-7B-300B with Docker Model Runner:
```
docker model run hf.co/OpenNLPLab/TransNormerLLM2-7B-300B
```

Finished training?

by Yhyu13 - opened Feb 26, 2024

Discussion

Yhyu13

Feb 26, 2024

I'm just a crious community member, no researcher.

Is 0.3 trillion pre-trained token a finish training? Compared to what v1 has trined on, which is 1.4 T token.

Since 0.3 T already achieved comparable results to previous version, is it worth further training? Or, are you aiming for scaling up to larger model sizes?

Thanks!

OpenNLPLab

Owner Feb 27, 2024

Thank you for your interest in our model! The TNL2-7B-300B model is currently in the testing phase, so we have paused its training for the time being. Meanwhile, we are actively training the TNL3-15B model. Please feel free to check this link for more information: https://huggingface.co/OpenNLPLab/TransNormerLLM3-15B-Intermediate-Checkpoints.

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment