Instructions to use thatupiso/SmolLM2-FT-DPO2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use thatupiso/SmolLM2-FT-DPO2 with Transformers:

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

pipe = pipeline("text-generation", model="thatupiso/SmolLM2-FT-DPO2")
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("thatupiso/SmolLM2-FT-DPO2")
model = AutoModelForCausalLM.from_pretrained("thatupiso/SmolLM2-FT-DPO2", 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 thatupiso/SmolLM2-FT-DPO2 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "thatupiso/SmolLM2-FT-DPO2"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "thatupiso/SmolLM2-FT-DPO2",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/thatupiso/SmolLM2-FT-DPO2

SGLang

How to use thatupiso/SmolLM2-FT-DPO2 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 "thatupiso/SmolLM2-FT-DPO2" \
    --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": "thatupiso/SmolLM2-FT-DPO2",
		"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 "thatupiso/SmolLM2-FT-DPO2" \
        --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": "thatupiso/SmolLM2-FT-DPO2",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use thatupiso/SmolLM2-FT-DPO2 with Docker Model Runner:
```
docker model run hf.co/thatupiso/SmolLM2-FT-DPO2
```

thatupiso commited on Dec 12, 2024

Commit

82a3989

verified ·

1 Parent(s): 64ac2f5

End of training

Browse files

Files changed (5) hide show

README.md +2 -3
runs/Dec12_20-07-50_3393362a4d02/events.out.tfevents.1734034093.3393362a4d02.7235.0 +3 -0
runs/Dec12_20-23-09_3393362a4d02/events.out.tfevents.1734035019.3393362a4d02.7235.1 +3 -0
runs/Dec12_20-24-40_3393362a4d02/events.out.tfevents.1734035190.3393362a4d02.7235.2 +3 -0
training_args.bin +1 -1

README.md CHANGED Viewed

@@ -4,8 +4,7 @@ library_name: transformers
 model_name: SmolLM2-FT-DPO2
 tags:
 - generated_from_trainer
-- smol-course
-- module_1
 - trl
 - dpo
 licence: license
@@ -29,7 +28,7 @@ print(output["generated_text"])
 ## Training procedure
-[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/thatupiso-code-org/huggingface/runs/qr19ujp2)
 This model was trained with DPO, a method introduced in [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://huggingface.co/papers/2305.18290).

 model_name: SmolLM2-FT-DPO2
 tags:
 - generated_from_trainer
+- dpo-smolK12-100
 - trl
 - dpo
 licence: license
 ## Training procedure
+[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/thatupiso-code-org/huggingface/runs/xpcn3ywm)
 This model was trained with DPO, a method introduced in [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://huggingface.co/papers/2305.18290).

runs/Dec12_20-07-50_3393362a4d02/events.out.tfevents.1734034093.3393362a4d02.7235.0 ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:c4fe8439c4d0e4bf0e674877c5e8b1f87e3d1eca8060c9c4efde206930a3d294
+size 13260

runs/Dec12_20-23-09_3393362a4d02/events.out.tfevents.1734035019.3393362a4d02.7235.1 ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:0f9f0e0fe7b9f89ae5720ca6baf57453966f4c8c3158a921f02bb12b8d8aa3bf
+size 27116

runs/Dec12_20-24-40_3393362a4d02/events.out.tfevents.1734035190.3393362a4d02.7235.2 ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:853133961a02f0d971d003a31cb5cef46086c39367d37c950dd84c673478a3d0
+size 40308

training_args.bin CHANGED Viewed

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a1c3b48e802c46b7298aea7b9c84993251d389226ee3bef09e82528f7370ec07
 size 6072

 version https://git-lfs.github.com/spec/v1
+oid sha256:82c7343ca0657002fe93055a64e96268492c8ea1017027804786a56752e20079
 size 6072