Instructions to use google/gemma-2b with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use google/gemma-2b with Transformers:

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

pipe = pipeline("text-generation", model="google/gemma-2b")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("google/gemma-2b")
model = AutoModelForCausalLM.from_pretrained("google/gemma-2b")

llama-cpp-python

How to use google/gemma-2b with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="google/gemma-2b",
	filename="gemma-2b.gguf",
)

output = llm(
	"Once upon a time,",
	max_tokens=512,
	echo=True
)
print(output)

Inference
Notebooks
Google Colab
Kaggle
Local Apps

llama.cpp

How to use google/gemma-2b with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf google/gemma-2b
# Run inference directly in the terminal:
llama-cli -hf google/gemma-2b

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf google/gemma-2b
# Run inference directly in the terminal:
llama-cli -hf google/gemma-2b

Use pre-built binary

# Download pre-built binary from:
# https://github.com/ggerganov/llama.cpp/releases
# Start a local OpenAI-compatible server with a web UI:
./llama-server -hf google/gemma-2b
# Run inference directly in the terminal:
./llama-cli -hf google/gemma-2b

Build from source code

git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-server llama-cli
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama-server -hf google/gemma-2b
# Run inference directly in the terminal:
./build/bin/llama-cli -hf google/gemma-2b

Use Docker

docker model run hf.co/google/gemma-2b

LM Studio
Jan

vLLM

How to use google/gemma-2b with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "google/gemma-2b"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "google/gemma-2b",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/google/gemma-2b

SGLang

How to use google/gemma-2b 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 "google/gemma-2b" \
    --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": "google/gemma-2b",
		"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 "google/gemma-2b" \
        --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": "google/gemma-2b",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Ollama
How to use google/gemma-2b with Ollama:
```
ollama run hf.co/google/gemma-2b
```

Unsloth Studio new

How to use google/gemma-2b with Unsloth Studio:

Install Unsloth Studio (macOS, Linux, WSL)

curl -fsSL https://unsloth.ai/install.sh | sh
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for google/gemma-2b to start chatting

Install Unsloth Studio (Windows)

irm https://unsloth.ai/install.ps1 | iex
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for google/gemma-2b to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for google/gemma-2b to start chatting

Docker Model Runner
How to use google/gemma-2b with Docker Model Runner:
```
docker model run hf.co/google/gemma-2b
```

Lemonade

How to use google/gemma-2b with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull google/gemma-2b

Run and chat with the model

lemonade run user.gemma-2b-{{QUANT_TAG}}

List all available models

lemonade list

pretraining Gemma for domain dataset

#41

by Iamexperimenting - opened Mar 20, 2024

Discussion

Iamexperimenting

Mar 20, 2024

•

edited Mar 26, 2024

Hi team,

I would like to pretrain Gemma model with my domain dataset. I wanted to train gemma model with my domain data. But I wanted to train all the parameters not using LoRA.

1.a : does tokenizer learns/add any new tokens(domain specific words which is not present in the training tokenizer) during continued pre-training?

can you please provide any example article to fine-tune full parameters?

@ybelkada @suryabhupa

Iamexperimenting

Mar 25, 2024

@suryabhupa @ybelkada can you please provide an example?

suryabhupa

Google org Mar 26, 2024

Hello! Sorry for the delay.

I'm not sure what you mean by new tokens; you shouldn't need to use any new tokens when finetuning, and you are welcome to use any formatting template you'd like; see our own formatting we use if you'd like, especially as those are natively supported by our tokenizer.
The Zephyr 7B published their finetuning setup here: https://huggingface.co/HuggingFaceH4/zephyr-7b-gemma-v0.1, and other guides such as https://ai.google.dev/gemma/docs/jax_finetune, https://lightning.ai/lightning-ai/studios/understanding-using-and-finetuning-gemma, https://www.kaggle.com/code/lucamassaron/fine-tune-gemma-7b-it-for-sentiment-analysis exist as well.

Iamexperimenting changed discussion title from domain specific fine-tuning to pretraining Gemma for domain dataset Mar 26, 2024

Iamexperimenting

Mar 26, 2024

@suryabhupa @ybelkada
token which I was referring to "domain specific word(technical word) which is not present in the training tokenizer".

I think, I have used the wrong term in the title and above description. Now, I have changed it.

Basically, I wanted to train the Gemma model with my domain data(like I want to continue to train the Gemma model with my domain data). I wanted to train all parameters in the Gemma model.

Like Google team trained the base Gemma model. I would like to take the base Gemma model and continue to train it with my domain data.

Did you use delimiter <bos> and <eos> during training the base Gemma model?

Iamexperimenting

Apr 3, 2024

@suryabhupa @ybelkada can you please guide here?

Iamexperimenting

Apr 4, 2024

@suryabhupa @ybelkada can you please guide here?

ybelkada

Apr 10, 2024

Hi @Iamexperimenting
Thanks ! I will let @suryabhupa reply here whenever he can as I am not familiar with the Gemma training procedure

suryabhupa

Google org Apr 10, 2024

Hello! Yes, when constructing batches, I'd recommend having sequences in your pretraining set-up that have and tokens in the right places to delimit sequences, but also to properly construct the attention masks. We also use and tokens when doing so. You should experiment with how exactly you pack your examples into a single batch, I'd recommend checking out the T5, GPT, or PaLM paper for some details on how they did it. You do not need to add any extra tokens here.

sitandon

Jun 1, 2024

@suryabhupa : Based on my understanding GPT 2 inserts <|endoftext|> between documents and then chunk it to have same size (Please correct me here if otherwise). Curious to know how <bos> and <eos> has been set up for Gemma. Lets say seq_len is 3 and we have two documents:

Hello India
Hello how are you doing

Will it be done like this

First add special tokens to entire text corpus like this: <bos> Hello India <eos> <bos> Hello how are you doing <eos>
Split it based on seq len :
- <bos> Hello India
- <eos> <bos> Hello
- how are you
- doing <eos>

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