Instructions to use WilliamHH/Assignment2-modified-V4 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use WilliamHH/Assignment2-modified-V4 with Transformers:

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

pipe = pipeline("text-generation", model="WilliamHH/Assignment2-modified-V4")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("WilliamHH/Assignment2-modified-V4")
model = AutoModelForCausalLM.from_pretrained("WilliamHH/Assignment2-modified-V4")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use WilliamHH/Assignment2-modified-V4 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "WilliamHH/Assignment2-modified-V4"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "WilliamHH/Assignment2-modified-V4",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/WilliamHH/Assignment2-modified-V4

SGLang

How to use WilliamHH/Assignment2-modified-V4 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 "WilliamHH/Assignment2-modified-V4" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "WilliamHH/Assignment2-modified-V4",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

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 "WilliamHH/Assignment2-modified-V4" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "WilliamHH/Assignment2-modified-V4",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use WilliamHH/Assignment2-modified-V4 with Docker Model Runner:
```
docker model run hf.co/WilliamHH/Assignment2-modified-V4
```

Assignment2-modified-V4

This model is a fine-tuned version of HuggingFaceTB/SmolLM-135M on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 2.8664

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: 4e-05
train_batch_size: 8
eval_batch_size: 8
seed: 42
optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: cosine
num_epochs: 3

Training results

Training Loss	Epoch	Step	Validation Loss
2.5124	0.1067	200	3.1406
2.646	0.2133	400	3.0470
2.4984	0.32	600	2.9933
2.4831	0.4267	800	2.9577
2.4422	0.5333	1000	2.9183
2.3946	0.64	1200	2.8964
2.3091	0.7467	1400	2.8777
2.3445	0.8533	1600	2.8548
2.3067	0.96	1800	2.8339
2.1795	1.0667	2000	2.8570
1.9869	1.1733	2200	2.8557
1.9955	1.28	2400	2.8516
2.0386	1.3867	2600	2.8442
1.9987	1.4933	2800	2.8436
2.0111	1.6	3000	2.8368
1.9452	1.7067	3200	2.8283
1.9509	1.8133	3400	2.8191
1.9163	1.92	3600	2.8148
1.9621	2.0267	3800	2.8338
1.786	2.1333	4000	2.8653
1.7842	2.24	4200	2.8693
1.8084	2.3467	4400	2.8705
1.7057	2.4533	4600	2.8779
1.8589	2.56	4800	2.8660
1.7983	2.6667	5000	2.8655
1.7421	2.7733	5200	2.8659
1.7648	2.88	5400	2.8664
1.8171	2.9867	5600	2.8664

Framework versions

Transformers 4.57.1
Pytorch 2.8.0+cu126
Datasets 4.0.0
Tokenizers 0.22.1

Downloads last month: 12

Safetensors

Model size

0.1B params

Tensor type

F32

Model tree for WilliamHH/Assignment2-modified-V4

Base model

HuggingFaceTB/SmolLM-135M

Finetuned

(115)

this model