# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM
tokenizer = AutoTokenizer.from_pretrained("Langboat/bloom-6b4-zh")
model = AutoModelForCausalLM.from_pretrained("Langboat/bloom-6b4-zh")Quick Links
This model is based on bigscience/bloom-7b1.
We pruned its vocabulary from 250880 to 46145 with Chinese corpus to reduce GPU memory usage. So the total parameter is 6b4 now.
How to use
from transformers import BloomTokenizerFast, BloomForCausalLM
tokenizer = BloomTokenizerFast.from_pretrained('Langboat/bloom-6b4-zh')
model = BloomForCausalLM.from_pretrained('Langboat/bloom-6b4-zh')
print(tokenizer.batch_decode(model.generate(tokenizer.encode('中国的首都是', return_tensors='pt'))))
- Downloads last month
- 6
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="Langboat/bloom-6b4-zh")