| license: bigscience-bloom-rail-1.0 | |
| language: | |
| - en | |
| pipeline_tag: text-generation | |
| tags: | |
| - nlp | |
| - code | |
| ## Model Summary | |
| bloomz-3b-mmlu-lora is a LORA model which fine-tuned on mmlu dataset. The base model is `bigscience/bloomz-3b` . | |
| ## How to Use | |
| ```python | |
| import torch | |
| from transformers import AutoTokenizer | |
| from peft import AutoPeftModelForCausalLM | |
| torch.set_default_device("cuda") | |
| model = AutoPeftModelForCausalLM.from_pretrained("liuchanghf/bloomz-3b-mmlu-lora") | |
| tokenizer = AutoTokenizer.from_pretrained("bigscience/bloomz-3b") | |
| inputs = tokenizer('''def print_prime(n): | |
| """ | |
| Print all primes between 1 and n | |
| """''', return_tensors="pt", return_attention_mask=False) | |
| outputs = model.generate(**inputs, max_length=200) | |
| text = tokenizer.batch_decode(outputs)[0] | |
| print(text) | |
| ``` |