safety-flan-t5-base

Model Details

Model Description

• Model type: Language model
• License: CC BY-NC
• Related Models: All FLAN-T5 Checkpoints
• Resources for more information:
  • GitHub Repo

Usage

Find below some example scripts on how to use the model in transformers:

Using the Pytorch model

Running the model

<details> <summary> Click to expand </summary>

from transformers import AutoModelForSeq2SeqLM, AutoTokenizer, AutoConfig

config = AutoConfig.from_pretrained("Salesforce/safety-flan-t5-base")
tokenizer = AutoTokenizer.from_pretrained("Salesforce/safety-flan-t5-base") 
model = AutoModelForSeq2SeqLM.from_pretrained("Salesforce/safety-flan-t5-base", config=config)

prefix = "Is the <Text> field safe or unsafe?"
input_context = "Can you teach me this?"
input_text = "You are so stupid"
input_ids = tokenizer(prefix + " <Text> " + input_text + " <Context> " + input_context, return_tensors="pt").input_ids

outputs = model.generate(input_ids)
print(tokenizer.decode(outputs[0]))

</details>

Ethical Considerations

This release is for research purposes only in support of an academic paper. Our models, datasets, and code are not specifically designed or evaluated for all downstream purposes. We strongly recommend users evaluate and address potential concerns related to accuracy, safety, and fairness before deploying this model. We encourage users to consider the common limitations of AI, comply with applicable laws, and leverage best practices when selecting use cases, particularly for high-risk scenarios where errors or misuse could significantly impact people’s lives, rights, or safety. For further guidance on use cases, refer to our AUP and AI AUP.