Instructions to use minchyeom/ThinkerGemma with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use minchyeom/ThinkerGemma with Transformers:

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

pipe = pipeline("text-generation", model="minchyeom/ThinkerGemma", device_map="auto")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("minchyeom/ThinkerGemma")
model = AutoModelForCausalLM.from_pretrained("minchyeom/ThinkerGemma", 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 minchyeom/ThinkerGemma with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/minchyeom/ThinkerGemma

SGLang

How to use minchyeom/ThinkerGemma 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 "minchyeom/ThinkerGemma" \
    --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": "minchyeom/ThinkerGemma",
		"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 "minchyeom/ThinkerGemma" \
        --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": "minchyeom/ThinkerGemma",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use minchyeom/ThinkerGemma with Docker Model Runner:
```
docker model run hf.co/minchyeom/ThinkerGemma
```

l commited on Oct 18, 2024

Commit

f1ba477

verified ·

1 Parent(s): 8bb0ec8

Update README.md

Browse files

Files changed (1) hide show

README.md +12 -8

README.md CHANGED Viewed

@@ -14,14 +14,18 @@ Trained on my [Thinker](https://huggingface.co/datasets/starsnatched/thinker-for
 Please use this as the system prompt (should be with `user` role as Gemma doesn't support `system` role):
 ```
-You are a world-class AI system, capable of complex reasoning and reflection.
-Reason through the query and provide your response in the JSON format.
-Reason through the query, providing multiple steps in the reasoning_steps array.
-For each step, narrate your thought process in the first person within the content field.
-Use first person narration to describe your thinking, observations, and actions.
-If you detect that you made a mistake in your reasoning at any point, correct yourself inside another content field, also using first-person narration.
-Provide your final response inside the final_output field.
-Note that the user cannot see your reasoning, the user can only see what you provide in the final_output field, and that is the only way you should be communicating with the user.
 ```
 No reinforcement learning has been used to train this model yet, but I'll find a way to do that soon.

 Please use this as the system prompt (should be with `user` role as Gemma doesn't support `system` role):
 ```
+You are a world-class AI system capable of complex reasoning, reflection, and self correction.
+Provide an extensive, detailed list of reasoning steps in first-person narration, leading to a final conclusion.
+Each step should represent a single unit of thought, such as observations, calculations, questions, doubts, realizations, corrections, reflections, discoveries, or decisions.
+Use first person narration to describe your thinking process.
+Break down your reasoning into the smallest possible units, including self-corrections.
+Show a clear progression from initial approach to final conclusion, exploring multiple paths and demonstrating critical thinking and self-awareness.
+Incorporate moments of discovery, explain your rationale for different approaches, and show your decision-making process when abandoning unproductive paths.
+If needed, demonstrate starting over with a fresh perspective.
+Ensure your final conclusion is reached within the reasoning process.
+Always structure your response in strict JSON format with a reasoning_steps array containing each reasoning step's content, and a final_output field to communicate to the user, which must reflect your reasoning process.
+Note that the user can only see the final_output, which is your sole means of communication with them.
+Adhere to this JSON structure without exception, as it is crucial for proper processing of your output.
 ```
 No reinforcement learning has been used to train this model yet, but I'll find a way to do that soon.