Instructions to use zehao888/gpt2-pretrain_new with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use zehao888/gpt2-pretrain_new with Transformers:

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

pipe = pipeline("text-generation", model="zehao888/gpt2-pretrain_new")

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("zehao888/gpt2-pretrain_new")
model = AutoModelForCausalLM.from_pretrained("zehao888/gpt2-pretrain_new", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use zehao888/gpt2-pretrain_new with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "zehao888/gpt2-pretrain_new"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "zehao888/gpt2-pretrain_new",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/zehao888/gpt2-pretrain_new

SGLang

How to use zehao888/gpt2-pretrain_new 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 "zehao888/gpt2-pretrain_new" \
    --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": "zehao888/gpt2-pretrain_new",
		"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 "zehao888/gpt2-pretrain_new" \
        --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": "zehao888/gpt2-pretrain_new",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use zehao888/gpt2-pretrain_new with Docker Model Runner:
```
docker model run hf.co/zehao888/gpt2-pretrain_new
```

gpt2-pretrain_new

This model is a fine-tuned version of gpt2 on the None dataset. It achieves the following results on the evaluation set:

Loss: 5.1443

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: 3e-05
train_batch_size: 48
eval_batch_size: 48
seed: 42
optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: cosine
lr_scheduler_warmup_ratio: 0.1
num_epochs: 50

Training results

Training Loss	Epoch	Step	Validation Loss
8.3786	1.8519	100	8.0241
7.5755	3.7037	200	7.1144
6.7409	5.5556	300	6.4557
6.171	7.4074	400	6.0812
5.8181	9.2593	500	5.8451
5.5475	11.1111	600	5.6727
5.3334	12.9630	700	5.5440
5.1484	14.8148	800	5.4523
4.9984	16.6667	900	5.3853
4.8624	18.5185	1000	5.3275
4.7532	20.3704	1100	5.2845
4.633	22.2222	1200	5.2454
4.5402	24.0741	1300	5.2166
4.444	25.9259	1400	5.1897
4.3621	27.7778	1500	5.1785
4.2848	29.6296	1600	5.1639
4.2343	31.4815	1700	5.1573
4.1692	33.3333	1800	5.1483
4.125	35.1852	1900	5.1528
4.087	37.0370	2000	5.1461
4.0493	38.8889	2100	5.1443
4.027	40.7407	2200	5.1476
4.0032	42.5926	2300	5.1464

Framework versions

Transformers 4.57.1
Pytorch 2.9.0+cu126
Datasets 4.0.0
Tokenizers 0.22.1

Downloads last month: 7

Safetensors

Model size

0.1B params

Tensor type

F32

Model tree for zehao888/gpt2-pretrain_new

Base model

openai-community/gpt2

Finetuned

(2301)

this model

Finetunes

1 model