Instructions to use google/flan-t5-xl with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use google/flan-t5-xl with Transformers:
# Load model directly from transformers import AutoTokenizer, AutoModelForSeq2SeqLM tokenizer = AutoTokenizer.from_pretrained("google/flan-t5-xl") model = AutoModelForSeq2SeqLM.from_pretrained("google/flan-t5-xl") - Notebooks
- Google Colab
- Kaggle
can anyone help me to get Prompt template for question answering model
Hi Team,
Currently, I'm working on building "Question answering model using opensource LLM". Currently, I'm using the below template to generate answering
"""
{context}\n\n{question}
"""
This template provides me one word answer.
Example
Question : what is the maximum kilometer did Andrew covered during his cycling practice?
Answer: 28 Kilometer
Is it possible to generate answers like conversational ai(human-like reply)
example
Question : what is the maximum kilometer did Andrew covered during his cycling practice?
Answer: Maximum kilometer covered by Andrew is 28.
can anyone please help me here?
Hi @Iamexperimenting
You may want to try out different sampling techniques first: https://huggingface.co/blog/how-to-generate and see if this helps
For generating better responses in general I would recommend to also try out different models such as mistral-7b: https://huggingface.co/mistralai/Mistral-7B-v0.1 or Llama-2