Instructions to use goldfish-models/tzo_latn_full with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use goldfish-models/tzo_latn_full with Transformers:

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

pipe = pipeline("text-generation", model="goldfish-models/tzo_latn_full")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("goldfish-models/tzo_latn_full")
model = AutoModelForCausalLM.from_pretrained("goldfish-models/tzo_latn_full")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use goldfish-models/tzo_latn_full with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/goldfish-models/tzo_latn_full

SGLang

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

Docker Model Runner
How to use goldfish-models/tzo_latn_full with Docker Model Runner:
```
docker model run hf.co/goldfish-models/tzo_latn_full
```

tzo_latn_full / README.md

goldfish-models

Upload README.md with huggingface_hub

1b2b27c verified over 1 year ago

preview code

raw

history blame contribute delete

2.59 kB


	---
	license: apache-2.0
	language:
	- tzo
	datasets:
	- allenai/MADLAD-400
	library_name: transformers
	pipeline_tag: text-generation
	tags:
	- goldfish
	- arxiv:2408.10441
	---

	# tzo_latn_full

	Goldfish is a suite of monolingual language models trained for 350 languages.
	This model is the <b>Tzotzil</b> (Latin script) model trained on 9MB of data (all our data in the language), after accounting for an estimated byte premium of 1.49; content-matched text in Tzotzil takes on average 1.49x as many UTF-8 bytes to encode as English.
	The Goldfish models are trained primarily for comparability across languages and for low-resource languages; Goldfish performance for high-resource languages is not designed to be comparable with modern large language models (LLMs).

	Note: tzo_latn is an [individual language](https://iso639-3.sil.org/code_tables/639/data) code. It is not contained in any macrolanguage codes contained in Goldfish (for script latn).

	All training and hyperparameter details are in our paper, [Goldfish: Monolingual Language Models for 350 Languages (Chang et al., 2024)](https://www.arxiv.org/abs/2408.10441).

	Training code and sample usage: https://github.com/tylerachang/goldfish

	Sample usage also in this Google Colab: [link](https://colab.research.google.com/drive/1rHFpnQsyXJ32ONwCosWZ7frjOYjbGCXG?usp=sharing)

	## Model details:

	To access all Goldfish model details programmatically, see https://github.com/tylerachang/goldfish/blob/main/model_details.json.
	All models are trained with a [CLS] (same as [BOS]) token prepended, and a [SEP] (same as [EOS]) token separating sequences.
	For best results, make sure that [CLS] is prepended to your input sequence (see sample usage linked above)!
	Details for this model specifically:

	* Architecture: gpt2
	* Parameters: 121625088
	* Maximum sequence length: 512 tokens
	* Training text data (raw): 13.42MB
	* Training text data (byte premium scaled): 9.025MB
	* Training tokens: 3463680 (x10 epochs)
	* Vocabulary size: 45078
	* Compute cost: 1.7668828495872e+16 FLOPs or ~1.7 NVIDIA A6000 GPU hours

	Training datasets (percentages prior to deduplication):
	* 50.76388%: [MADLAD-400 (CommonCrawl)](https://huggingface.co/datasets/allenai/MADLAD-400)
	* 34.89747%: [eBible](https://ebible.org/find/)


	## Citation

	If you use this model, please cite:

	```
	@article{chang-etal-2024-goldfish,
	title={Goldfish: Monolingual Language Models for 350 Languages},
	author={Chang, Tyler A. and Arnett, Catherine and Tu, Zhuowen and Bergen, Benjamin K.},
	journal={Preprint},
	year={2024},
	url={https://www.arxiv.org/abs/2408.10441},
	}
	```