This model is based on bigscience/bloom-560m.
We pruned its vocabulary from 250880 to 42437 with Chinese corpus to reduce GPU memory usage. So the total parameter is 389m now.
How to use
from transformers import BloomTokenizerFast, BloomForCausalLM
tokenizer = BloomTokenizerFast.from_pretrained('Langboat/bloom-389m-zh')
model = BloomForCausalLM.from_pretrained('Langboat/bloom-389m-zh')
print(tokenizer.batch_decode(model.generate(tokenizer.encode('中国的首都是', return_tensors='pt'))))
- Downloads last month
- 368
This model does not have enough activity to be deployed to Inference API (serverless) yet.
Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated)
instead.