Instructions to use Jasonnn13/SmolLM2-FT-MyDataset with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Jasonnn13/SmolLM2-FT-MyDataset with Transformers:

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

pipe = pipeline("text-generation", model="Jasonnn13/SmolLM2-FT-MyDataset")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("Jasonnn13/SmolLM2-FT-MyDataset")
model = AutoModelForCausalLM.from_pretrained("Jasonnn13/SmolLM2-FT-MyDataset")
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

vLLM

How to use Jasonnn13/SmolLM2-FT-MyDataset with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Jasonnn13/SmolLM2-FT-MyDataset"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Jasonnn13/SmolLM2-FT-MyDataset",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Jasonnn13/SmolLM2-FT-MyDataset

SGLang

How to use Jasonnn13/SmolLM2-FT-MyDataset 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 "Jasonnn13/SmolLM2-FT-MyDataset" \
    --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": "Jasonnn13/SmolLM2-FT-MyDataset",
		"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 "Jasonnn13/SmolLM2-FT-MyDataset" \
        --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": "Jasonnn13/SmolLM2-FT-MyDataset",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use Jasonnn13/SmolLM2-FT-MyDataset with Docker Model Runner:
```
docker model run hf.co/Jasonnn13/SmolLM2-FT-MyDataset
```

SmolLM2-FT-MyDataset

File size: 2,391 Bytes

36d744b

---
library_name: transformers
license: apache-2.0
base_model: HuggingFaceTB/SmolLM2-135M
tags:
- smol-course
- module_1
- trl
- sft
- trl
- sft
- generated_from_trainer
model-index:
- name: SmolLM2-FT-MyDataset
  results: []
---

<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->

# SmolLM2-FT-MyDataset

This model is a fine-tuned version of [HuggingFaceTB/SmolLM2-135M](https://huggingface.co/HuggingFaceTB/SmolLM2-135M) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 0.9266

## 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: 5e-05
- train_batch_size: 4
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- training_steps: 1000

### Training results

| Training Loss | Epoch  | Step | Validation Loss |
|:-------------:|:------:|:----:|:---------------:|
| 1.0161        | 0.0982 | 50   | 1.0429          |
| 0.9943        | 0.1965 | 100  | 1.0086          |
| 0.953         | 0.2947 | 150  | 0.9906          |
| 0.9561        | 0.3929 | 200  | 0.9802          |
| 0.9585        | 0.4912 | 250  | 0.9699          |
| 0.9394        | 0.5894 | 300  | 0.9597          |
| 0.9284        | 0.6876 | 350  | 0.9535          |
| 0.9237        | 0.7859 | 400  | 0.9462          |
| 0.8945        | 0.8841 | 450  | 0.9402          |
| 0.9043        | 0.9823 | 500  | 0.9348          |
| 0.7479        | 1.0806 | 550  | 0.9421          |
| 0.805         | 1.1788 | 600  | 0.9388          |
| 0.7359        | 1.2770 | 650  | 0.9382          |
| 0.7437        | 1.3752 | 700  | 0.9365          |
| 0.7778        | 1.4735 | 750  | 0.9336          |
| 0.7678        | 1.5717 | 800  | 0.9317          |
| 0.7452        | 1.6699 | 850  | 0.9301          |
| 0.7373        | 1.7682 | 900  | 0.9279          |
| 0.738         | 1.8664 | 950  | 0.9269          |
| 0.7513        | 1.9646 | 1000 | 0.9266          |


### Framework versions

- Transformers 4.44.2
- Pytorch 2.10.0+cu128
- Datasets 4.0.0
- Tokenizers 0.19.1