Instructions to use Huysun29/cbt-gemma2-9b-v2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Huysun29/cbt-gemma2-9b-v2 with Transformers:

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

pipe = pipeline("text-generation", model="Huysun29/cbt-gemma2-9b-v2")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("Huysun29/cbt-gemma2-9b-v2")
model = AutoModelForCausalLM.from_pretrained("Huysun29/cbt-gemma2-9b-v2")
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 Huysun29/cbt-gemma2-9b-v2 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Huysun29/cbt-gemma2-9b-v2"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Huysun29/cbt-gemma2-9b-v2",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Huysun29/cbt-gemma2-9b-v2

SGLang

How to use Huysun29/cbt-gemma2-9b-v2 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 "Huysun29/cbt-gemma2-9b-v2" \
    --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": "Huysun29/cbt-gemma2-9b-v2",
		"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 "Huysun29/cbt-gemma2-9b-v2" \
        --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": "Huysun29/cbt-gemma2-9b-v2",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use Huysun29/cbt-gemma2-9b-v2 with Docker Model Runner:
```
docker model run hf.co/Huysun29/cbt-gemma2-9b-v2
```

cbt-gemma2-9b

Fine-tune (LoRA→merged) của google/gemma-2-9b-it cho trợ lý tham vấn theo liệu pháp CBT. Model sinh JSON có cấu trúc gồm: risk_level · technique · rationale · plan · response · safety_action.

Huấn luyện

Dữ liệu: 26.000 mẫu hội thoại CBT (định dạng messages, output JSON 6 trường).
LoRA r=32/α=64, lr=5e-05, max_len=1536, bf16.
best eval_loss = 0.46450522541999817 (epoch 2.0).

⚠️ Giới hạn & an toàn

Đây là công cụ hỗ trợ học thuật, KHÔNG thay thế chuyên gia sức khỏe tâm thần, KHÔNG chẩn đoán, KHÔNG tư vấn thuốc. Khi có dấu hiệu khủng hoảng/tự hại, model được huấn luyện để ưu tiên an toàn và khuyến nghị liên hệ chuyên gia/dịch vụ khẩn cấp. Cần có chuyên gia rà soát (human-in-the-loop) trước khi dùng với người thật.

Downloads last month: 8

Safetensors

Model size

9B params

Tensor type

BF16

Model tree for Huysun29/cbt-gemma2-9b-v2

Base model

google/gemma-2-9b

Finetuned

google/gemma-2-9b-it

Adapter

(480)

this model