# Fine tune text generation model

**URL:** <https://discuss.huggingface.co/t/fine-tune-text-generation-model/69381>\
**Category:** Beginners\
**Created:** [January 16, 2024, 2:45pm UTC](https://discuss.huggingface.co/t/fine-tune-text-generation-model/69381 "2024-01-16T14:45:25Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![atrujill](https://sea2.discourse-cdn.com/hellohellohello/user_avatar/discuss.huggingface.co/atrujill/32/23039_2.png) [@atrujill](https://discuss.huggingface.co/u/atrujill)\
**Post date:** [January 16, 2024, 2:45pm UTC](https://discuss.huggingface.co/t/fine-tune-text-generation-model/69381/1 "2024-01-16T14:45:25Z")

</div>

Hello, is it possible to fine tune a text generation model? I would like to generate an instance of a model to incorporate a dataset like the following: [UCI Machine Learning Repository](https://archive.ics.uci.edu/dataset/551/gas+turbine+co+and+nox+emission+data+set)

It is a dataset that has numerical values of operation of a natural gas turbine that generates electricity.

I wanto to use the model to make inferences such as asking: What would be the CO emission if the power is XX, the ambient temperature is YY, etc.

I know that it is possible to add information as context in the prompt, but they are long in text.

I dont know if the wrigth way is to make an adapter

I was able to generate a jsonl file from that dataset.

Thanks for your time.