Instructions to use NeverLearn/gpt2-wikitext2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use NeverLearn/gpt2-wikitext2 with Transformers:

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

pipe = pipeline("text-generation", model="NeverLearn/gpt2-wikitext2")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("NeverLearn/gpt2-wikitext2")
model = AutoModelForCausalLM.from_pretrained("NeverLearn/gpt2-wikitext2", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use NeverLearn/gpt2-wikitext2 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "NeverLearn/gpt2-wikitext2"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "NeverLearn/gpt2-wikitext2",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/NeverLearn/gpt2-wikitext2

SGLang

How to use NeverLearn/gpt2-wikitext2 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 "NeverLearn/gpt2-wikitext2" \
    --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": "NeverLearn/gpt2-wikitext2",
		"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 "NeverLearn/gpt2-wikitext2" \
        --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": "NeverLearn/gpt2-wikitext2",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use NeverLearn/gpt2-wikitext2 with Docker Model Runner:
```
docker model run hf.co/NeverLearn/gpt2-wikitext2
```

gpt2-wikitext2

This model is a fine-tuned version of openai-community/gpt2 on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 5.6547

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: 1e-05
train_batch_size: 8
eval_batch_size: 8
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
num_epochs: 60

Training results

Training Loss	Epoch	Step	Validation Loss
No log	1.0	55	9.1458
No log	2.0	110	8.3471
No log	3.0	165	7.7884
No log	4.0	220	7.3751
No log	5.0	275	7.0487
No log	6.0	330	6.7857
No log	7.0	385	6.5840
No log	8.0	440	6.4196
No log	9.0	495	6.2584
7.3272	10.0	550	6.1628
7.3272	11.0	605	6.0521
7.3272	12.0	660	5.9861
7.3272	13.0	715	5.9223
7.3272	14.0	770	5.8760
7.3272	15.0	825	5.8246
7.3272	16.0	880	5.7813
7.3272	17.0	935	5.7663
7.3272	18.0	990	5.7275
5.2638	19.0	1045	5.7022
5.2638	20.0	1100	5.6905
5.2638	21.0	1155	5.6803
5.2638	22.0	1210	5.6740
5.2638	23.0	1265	5.6631
5.2638	24.0	1320	5.6461
5.2638	25.0	1375	5.6326
5.2638	26.0	1430	5.6280
5.2638	27.0	1485	5.6408
4.5099	28.0	1540	5.6194
4.5099	29.0	1595	5.6255
4.5099	30.0	1650	5.6218
4.5099	31.0	1705	5.6127
4.5099	32.0	1760	5.6140
4.5099	33.0	1815	5.6281
4.5099	34.0	1870	5.6305
4.5099	35.0	1925	5.6139
4.5099	36.0	1980	5.6331
4.0571	37.0	2035	5.6323
4.0571	38.0	2090	5.6137
4.0571	39.0	2145	5.6258
4.0571	40.0	2200	5.6322
4.0571	41.0	2255	5.6392
4.0571	42.0	2310	5.6308
4.0571	43.0	2365	5.6329
4.0571	44.0	2420	5.6373
4.0571	45.0	2475	5.6407
3.7638	46.0	2530	5.6489
3.7638	47.0	2585	5.6489
3.7638	48.0	2640	5.6445
3.7638	49.0	2695	5.6428
3.7638	50.0	2750	5.6425
3.7638	51.0	2805	5.6450
3.7638	52.0	2860	5.6566
3.7638	53.0	2915	5.6504
3.7638	54.0	2970	5.6494
3.5759	55.0	3025	5.6538
3.5759	56.0	3080	5.6555
3.5759	57.0	3135	5.6529
3.5759	58.0	3190	5.6567
3.5759	59.0	3245	5.6551
3.5759	60.0	3300	5.6547

Framework versions

Transformers 4.40.0
Pytorch 2.2.1
Datasets 2.19.0
Tokenizers 0.19.1

Downloads last month: 6

Safetensors

Model size

0.1B params

Tensor type

F32

Model tree for NeverLearn/gpt2-wikitext2

Base model

openai-community/gpt2

Finetuned

(2252)

this model