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

Libraries

How to use system-k/gemma-2-2b-think with Transformers:

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

pipe = pipeline("text-generation", model="system-k/gemma-2-2b-think")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("system-k/gemma-2-2b-think")
model = AutoModelForCausalLM.from_pretrained("system-k/gemma-2-2b-think", device_map="auto")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use system-k/gemma-2-2b-think with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "system-k/gemma-2-2b-think"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "system-k/gemma-2-2b-think",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/system-k/gemma-2-2b-think

SGLang

How to use system-k/gemma-2-2b-think 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 "system-k/gemma-2-2b-think" \
    --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": "system-k/gemma-2-2b-think",
		"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 "system-k/gemma-2-2b-think" \
        --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": "system-k/gemma-2-2b-think",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

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

You need to agree to share your contact information to access this model

This repository is publicly accessible, but you have to accept the conditions to access its files and content.

sft

This model is a fine-tuned version of google/gemma-2-2b-it on the sk_sky_t1 dataset. It achieves the following results on the evaluation set:

Loss: 0.3965

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: 1e-05
train_batch_size: 1
eval_batch_size: 1
seed: 42
gradient_accumulation_steps: 2
total_train_batch_size: 2
optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: cosine
lr_scheduler_warmup_ratio: 0.1
num_epochs: 1.0

Training results

Training Loss	Epoch	Step	Validation Loss
0.6055	0.0335	500	0.5799
0.6442	0.0669	1000	0.5864
0.5074	0.1004	1500	0.6054
0.6184	0.1339	2000	0.5991
0.6338	0.1674	2500	0.5784
0.6266	0.2008	3000	0.5759
0.4934	0.2343	3500	0.5649
0.5326	0.2678	4000	0.5536
0.5324	0.3012	4500	0.5437
0.6128	0.3347	5000	0.5311
0.6001	0.3682	5500	0.5204
0.5169	0.4016	6000	0.5109
0.5012	0.4351	6500	0.5014
0.5338	0.4686	7000	0.4935
0.5349	0.5021	7500	0.4793
0.5438	0.5355	8000	0.4699
0.4265	0.5690	8500	0.4599
0.4636	0.6025	9000	0.4497
0.481	0.6359	9500	0.4415
0.5104	0.6694	10000	0.4336
0.4597	0.7029	10500	0.4261
0.483	0.7364	11000	0.4191
0.417	0.7698	11500	0.4121
0.4537	0.8033	12000	0.4075
0.3964	0.8368	12500	0.4028
0.4793	0.8702	13000	0.3999
0.489	0.9037	13500	0.3980
0.3901	0.9372	14000	0.3970
0.4054	0.9706	14500	0.3965

Framework versions

Transformers 4.46.1
Pytorch 2.5.1+cu124
Datasets 3.1.0
Tokenizers 0.20.3

Downloads last month: -

Safetensors

Model size

3B params

Tensor type

F32

Model tree for system-k/gemma-2-2b-think

Base model

google/gemma-2-2b

Finetuned

google/gemma-2-2b-it

Finetuned

(1099)

this model