Instructions to use KirubaLS/gemma2b-merged1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use KirubaLS/gemma2b-merged1 with PEFT:

from peft import PeftModel
from transformers import AutoModelForCausalLM

base_model = AutoModelForCausalLM.from_pretrained("google/gemma-2b")
model = PeftModel.from_pretrained(base_model, "KirubaLS/gemma2b-merged1")

Transformers

How to use KirubaLS/gemma2b-merged1 with Transformers:

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

pipe = pipeline("text-generation", model="KirubaLS/gemma2b-merged1")

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("KirubaLS/gemma2b-merged1")
model = AutoModelForCausalLM.from_pretrained("KirubaLS/gemma2b-merged1", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use KirubaLS/gemma2b-merged1 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "KirubaLS/gemma2b-merged1"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "KirubaLS/gemma2b-merged1",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/KirubaLS/gemma2b-merged1

SGLang

How to use KirubaLS/gemma2b-merged1 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 "KirubaLS/gemma2b-merged1" \
    --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": "KirubaLS/gemma2b-merged1",
		"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 "KirubaLS/gemma2b-merged1" \
        --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": "KirubaLS/gemma2b-merged1",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use KirubaLS/gemma2b-merged1 with Docker Model Runner:
```
docker model run hf.co/KirubaLS/gemma2b-merged1
```

gemma2b-merged1

This model is a fine-tuned version of google/gemma-2b on the None dataset. It achieves the following results on the evaluation set:

Loss: 1.5603

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: 0.0002
train_batch_size: 2
eval_batch_size: 2
seed: 42
gradient_accumulation_steps: 4
total_train_batch_size: 8
optimizer: Use OptimizerNames.PAGED_ADAMW_8BIT with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: linear
num_epochs: 5
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
1.1739	1.0	86	1.1135
0.9078	2.0	172	1.1035
0.7077	3.0	258	1.1719
0.4307	4.0	344	1.3608
0.3097	5.0	430	1.5603

Framework versions

PEFT 0.16.0
Transformers 4.54.1
Pytorch 2.6.0+cu124
Datasets 4.0.0
Tokenizers 0.21.4

Downloads last month: 10

Safetensors

Model size

3B params

Tensor type

F16

Model tree for KirubaLS/gemma2b-merged1

Base model

google/gemma-2b

Adapter

(23700)

this model