Instructions to use davron04/gemma-3-270m-uzen-base with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use davron04/gemma-3-270m-uzen-base with Transformers:

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

pipe = pipeline("text-generation", model="davron04/gemma-3-270m-uzen-base")

# Load model directly
from transformers import AutoTokenizer, AutoModelForMultimodalLM

tokenizer = AutoTokenizer.from_pretrained("davron04/gemma-3-270m-uzen-base")
model = AutoModelForMultimodalLM.from_pretrained("davron04/gemma-3-270m-uzen-base")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use davron04/gemma-3-270m-uzen-base with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "davron04/gemma-3-270m-uzen-base"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "davron04/gemma-3-270m-uzen-base",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/davron04/gemma-3-270m-uzen-base

SGLang

How to use davron04/gemma-3-270m-uzen-base 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 "davron04/gemma-3-270m-uzen-base" \
    --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": "davron04/gemma-3-270m-uzen-base",
		"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 "davron04/gemma-3-270m-uzen-base" \
        --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": "davron04/gemma-3-270m-uzen-base",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use davron04/gemma-3-270m-uzen-base with Docker Model Runner:
```
docker model run hf.co/davron04/gemma-3-270m-uzen-base
```

gemma-3-270m-uzen-base

This model is a fine-tuned version of davron04/gemma-3-270m-uzen-base on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 2.1987
Perplexity: 9.0416

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

learning_rate: 2e-05
train_batch_size: 2
eval_batch_size: 4
seed: 42
distributed_type: multi-GPU
num_devices: 2
gradient_accumulation_steps: 64
total_train_batch_size: 256
total_eval_batch_size: 8
optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: inverse_sqrt
lr_scheduler_warmup_steps: 0.01
num_epochs: 1
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Perplexity
No log	0	0	3.3726	29.0075
152.2381	0.1002	428	2.4411	11.5116
145.3987	0.2003	856	2.3491	10.5032
143.7446	0.3005	1284	2.3286	10.2931
140.7659	0.4006	1712	2.2912	9.9159
139.1574	0.5008	2140	2.2643	9.6535
137.4137	0.6009	2568	2.2431	9.4512
136.3983	0.7011	2996	2.2254	9.2859
135.6059	0.8012	3424	2.2111	9.1541
134.8424	0.9014	3852	2.1987	9.0416

Framework versions

Transformers 5.0.0
Pytorch 2.10.0+cu128
Datasets 4.8.5
Tokenizers 0.22.2

Downloads last month: 122

Safetensors

Model size

0.3B params

Tensor type

F32

Model tree for davron04/gemma-3-270m-uzen-base

Unable to build the model tree, the base model loops to the model itself. Learn more.