Instructions to use xiulinyang/GPT2_BABYLM10MRR_50000 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use xiulinyang/GPT2_BABYLM10MRR_50000 with Transformers:

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

pipe = pipeline("text-generation", model="xiulinyang/GPT2_BABYLM10MRR_50000")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("xiulinyang/GPT2_BABYLM10MRR_50000")
model = AutoModelForCausalLM.from_pretrained("xiulinyang/GPT2_BABYLM10MRR_50000", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use xiulinyang/GPT2_BABYLM10MRR_50000 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "xiulinyang/GPT2_BABYLM10MRR_50000"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "xiulinyang/GPT2_BABYLM10MRR_50000",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/xiulinyang/GPT2_BABYLM10MRR_50000

SGLang

How to use xiulinyang/GPT2_BABYLM10MRR_50000 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 "xiulinyang/GPT2_BABYLM10MRR_50000" \
    --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": "xiulinyang/GPT2_BABYLM10MRR_50000",
		"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 "xiulinyang/GPT2_BABYLM10MRR_50000" \
        --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": "xiulinyang/GPT2_BABYLM10MRR_50000",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use xiulinyang/GPT2_BABYLM10MRR_50000 with Docker Model Runner:
```
docker model run hf.co/xiulinyang/GPT2_BABYLM10MRR_50000
```

10MRR_50000_41

This model was trained from scratch on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 4.6435
Accuracy: 0.3132

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: 0.0006
train_batch_size: 32
eval_batch_size: 32
seed: 41
optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: linear
lr_scheduler_warmup_steps: 2000
num_epochs: 20.0
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.3222	1.0	1016	5.0527	0.2888
1.1829	2.0	2032	4.9731	0.2931
1.1528	3.0	3048	4.8485	0.2999
1.128	4.0	4064	4.7903	0.3020
1.1098	5.0	5080	4.7537	0.3039
1.0951	6.0	6096	4.7167	0.3058
1.0824	7.0	7112	4.6997	0.3068
1.0724	8.0	8128	4.6741	0.3081
1.0626	9.0	9144	4.6690	0.3082
1.0532	10.0	10160	4.6595	0.3097
1.0428	11.0	11176	4.6561	0.3101
1.0351	12.0	12192	4.6442	0.3108
1.0274	13.0	13208	4.6444	0.3109
1.0206	14.0	14224	4.6422	0.3117
1.0125	15.0	15240	4.6393	0.3118
1.0029	16.0	16256	4.6410	0.3124
0.9949	17.0	17272	4.6424	0.3123
0.9891	18.0	18288	4.6434	0.3128
0.9799	19.0	19304	4.6415	0.3130
0.977	20.0	20320	4.6435	0.3132

Framework versions

Transformers 4.51.3
Pytorch 2.5.1+cu121
Datasets 3.6.0
Tokenizers 0.21.1

Downloads last month: 6

Safetensors

Model size

0.1B params

Tensor type

F32