Instructions to use fausap/peft-smollm2-lora-gtx1660 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use fausap/peft-smollm2-lora-gtx1660 with PEFT:

from peft import PeftModel
from transformers import AutoModelForCausalLM

base_model = AutoModelForCausalLM.from_pretrained("HuggingFaceTB/SmolLM2-360M")
model = PeftModel.from_pretrained(base_model, "fausap/peft-smollm2-lora-gtx1660")

Transformers

How to use fausap/peft-smollm2-lora-gtx1660 with Transformers:

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

pipe = pipeline("text-generation", model="fausap/peft-smollm2-lora-gtx1660")

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("fausap/peft-smollm2-lora-gtx1660", dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use fausap/peft-smollm2-lora-gtx1660 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "fausap/peft-smollm2-lora-gtx1660"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "fausap/peft-smollm2-lora-gtx1660",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/fausap/peft-smollm2-lora-gtx1660

SGLang

How to use fausap/peft-smollm2-lora-gtx1660 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 "fausap/peft-smollm2-lora-gtx1660" \
    --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": "fausap/peft-smollm2-lora-gtx1660",
		"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 "fausap/peft-smollm2-lora-gtx1660" \
        --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": "fausap/peft-smollm2-lora-gtx1660",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use fausap/peft-smollm2-lora-gtx1660 with Docker Model Runner:
```
docker model run hf.co/fausap/peft-smollm2-lora-gtx1660
```

peft-smollm2-lora-gtx1660

This model is a fine-tuned version of HuggingFaceTB/SmolLM2-360M on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 3.6778

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: 0.0005
train_batch_size: 1
eval_batch_size: 1
seed: 42
gradient_accumulation_steps: 16
total_train_batch_size: 16
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
lr_scheduler_warmup_steps: 5
training_steps: 500

Training results

Training Loss	Epoch	Step	Validation Loss
4.0994	0.02	10	4.0259
3.9631	0.04	20	3.8910
3.915	0.06	30	3.8351
3.8301	0.08	40	3.7982
3.813	0.1	50	3.7773
3.7831	0.12	60	3.7633
3.7447	0.14	70	3.7478
3.7448	0.16	80	3.7437
3.7424	0.18	90	3.7297
3.7015	0.2	100	3.7205
3.7006	0.22	110	3.7144
3.6684	0.24	120	3.7020
3.6689	0.26	130	3.6980
3.6341	0.28	140	3.6918
3.6516	0.3	150	3.6897
3.6409	0.32	160	3.6922
3.6305	0.34	170	3.6829
3.617	0.36	180	3.6834
3.6111	0.38	190	3.6810
3.6092	0.4	200	3.6814
3.5892	0.42	210	3.6795
3.5968	0.44	220	3.6739
3.5732	0.46	230	3.6803
3.586	0.48	240	3.6729
3.5805	0.5	250	3.6765
3.5651	0.52	260	3.6788
3.5532	0.54	270	3.6749
3.556	0.56	280	3.6752
3.5717	0.58	290	3.6752
3.5333	0.6	300	3.6755
3.5652	0.62	310	3.6790
3.5473	0.64	320	3.6774
3.5352	0.66	330	3.6765
3.5369	0.68	340	3.6757
3.5356	0.7	350	3.6779
3.5418	0.72	360	3.6773
3.5458	0.74	370	3.6758
3.5502	0.76	380	3.6777
3.5114	0.78	390	3.6776
3.5532	0.8	400	3.6779
3.5411	0.82	410	3.6787
3.5357	0.84	420	3.6774
3.5353	0.86	430	3.6778
3.5408	0.88	440	3.6779
3.5562	0.9	450	3.6786
3.5272	0.92	460	3.6779
3.545	0.94	470	3.6776
3.5353	0.96	480	3.6776
3.5436	0.98	490	3.6778
3.5326	1.0	500	3.6778

Framework versions

PEFT 0.17.1
Transformers 4.57.1
Pytorch 2.8.0+cu128
Datasets 4.2.0
Tokenizers 0.22.1

Downloads last month: 15

Model tree for fausap/peft-smollm2-lora-gtx1660

Base model

HuggingFaceTB/SmolLM2-360M

Adapter

(16)

this model