Instructions to use Haleshot/Mathmate-7B-DELLA-ORPO-D with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Haleshot/Mathmate-7B-DELLA-ORPO-D with Transformers:

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

pipe = pipeline("text-generation", model="Haleshot/Mathmate-7B-DELLA-ORPO-D")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

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

tokenizer = AutoTokenizer.from_pretrained("Haleshot/Mathmate-7B-DELLA-ORPO-D")
model = AutoModelForCausalLM.from_pretrained("Haleshot/Mathmate-7B-DELLA-ORPO-D", 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=256)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use Haleshot/Mathmate-7B-DELLA-ORPO-D with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Haleshot/Mathmate-7B-DELLA-ORPO-D"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Haleshot/Mathmate-7B-DELLA-ORPO-D",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Haleshot/Mathmate-7B-DELLA-ORPO-D

SGLang

How to use Haleshot/Mathmate-7B-DELLA-ORPO-D 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 "Haleshot/Mathmate-7B-DELLA-ORPO-D" \
    --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": "Haleshot/Mathmate-7B-DELLA-ORPO-D",
		"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 "Haleshot/Mathmate-7B-DELLA-ORPO-D" \
        --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": "Haleshot/Mathmate-7B-DELLA-ORPO-D",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use Haleshot/Mathmate-7B-DELLA-ORPO-D with Docker Model Runner:
```
docker model run hf.co/Haleshot/Mathmate-7B-DELLA-ORPO-D
```

Mathmate-7B-DELLA-ORPO-D

Mathmate-7B-DELLA-ORPO-D is a finetuned version of Haleshot/Mathmate-7B-DELLA-ORPO using the ORPO method, combined with a LoRA adapter trained on everyday conversations.

Model Details

Base Model: Haleshot/Mathmate-7B-DELLA-ORPO
Training Dataset: HuggingFaceTB/everyday-conversations-llama3.1-2k

Dataset

The model incorporates training on the HuggingFaceTB/everyday-conversations-llama3.1-2k dataset, which focuses on everyday conversations and small talk.

Usage

Here's an example of how to use the model:

from transformers import AutoTokenizer, AutoModelForCausalLM
import torch

model_name = "Haleshot/Mathmate-7B-DELLA-ORPO-ORPO-D"
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForCausalLM.from_pretrained(model_name, torch_dtype=torch.float16, device_map="auto")

def generate_response(prompt, max_length=512):
    inputs = tokenizer(prompt, return_tensors="pt").to(model.device)
    outputs = model.generate(**inputs, max_length=max_length, num_return_sequences=1, do_sample=True, temperature=0.7)
    return tokenizer.decode(outputs[0], skip_special_tokens=True)

prompt = "Let's have a casual conversation about weekend plans."
response = generate_response(prompt)
print(response)

Acknowledgements

Thanks to the HuggingFaceTB team for providing the everyday conversations dataset used in this finetuning process.

Downloads last month: 13

Safetensors

Model size

7B params

Tensor type

F32

Model tree for Haleshot/Mathmate-7B-DELLA-ORPO-D

Base model

Haleshot/Mathmate-7B-DELLA

Finetuned

Haleshot/Mathmate-7B-DELLA-ORPO