Instructions to use yhavinga/gpt-neo-125M-dutch with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use yhavinga/gpt-neo-125M-dutch with Transformers:

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

pipe = pipeline("text-generation", model="yhavinga/gpt-neo-125M-dutch")

# Load model directly
from transformers import AutoTokenizer, AutoModelForMultimodalLM

tokenizer = AutoTokenizer.from_pretrained("yhavinga/gpt-neo-125M-dutch")
model = AutoModelForMultimodalLM.from_pretrained("yhavinga/gpt-neo-125M-dutch")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use yhavinga/gpt-neo-125M-dutch with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "yhavinga/gpt-neo-125M-dutch"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "yhavinga/gpt-neo-125M-dutch",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/yhavinga/gpt-neo-125M-dutch

SGLang

How to use yhavinga/gpt-neo-125M-dutch 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 "yhavinga/gpt-neo-125M-dutch" \
    --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": "yhavinga/gpt-neo-125M-dutch",
		"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 "yhavinga/gpt-neo-125M-dutch" \
        --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": "yhavinga/gpt-neo-125M-dutch",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use yhavinga/gpt-neo-125M-dutch with Docker Model Runner:
```
docker model run hf.co/yhavinga/gpt-neo-125M-dutch
```

gpt-neo-125M-dutch / run_gpt_neo.sh

yhavinga

Saving scripts, log and checkpoint at step 70000

f1818f3 over 4 years ago

raw

history blame

1.35 kB

	#!/bin/bash

	export HF_PROJECT="gpt-neo-125M-dutch-2"

	# Variables for training the tokenizer and creating the config
	export VOCAB_SIZE="50257"
	export DATASET="yhavinga/mc4_nl_cleaned" # Name of the dataset in the Huggingface Hub
	export DATASET_CONFIG="full" # Config of the dataset in the Huggingface Hub
	export DATASET_SPLIT="train" # Split to use for training tokenizer and model
	export TEXT_FIELD="text" # Field containing the text to be used for training
	export CONFIG_TYPE="EleutherAI/gpt-neo-125M" # Config that our model will use
	export MODEL_PATH="${HOME}/data/${HF_PROJECT}" # Path to the model, e.g. here inside the mount


	python run_clm_flax.py \
	--output_dir="${MODEL_PATH}" \
	--model_type="gpt_neo" \
	--config_name="${MODEL_PATH}" \
	--tokenizer_name="${MODEL_PATH}" \
	--preprocessing_num_workers="96" \
	--do_train --do_eval \
	--dataset_name="${DATASET}" \
	--dataset_config_name="${DATASET_CONFIG}" \
	--block_size="512" \
	--per_device_train_batch_size="16" \
	--per_device_eval_batch_size="16" \
	--learning_rate="0.0024" --warmup_steps="5000" \
	--adafactor \
	--overwrite_output_dir \
	--num_train_epochs="1" \
	--logging_steps="500" \
	--save_steps="10000" \
	--eval_steps="2500" \

	# --push_to_hub
	# --adam_beta1="0.9" --adam_beta2="0.98" --weight_decay="0.01" \