This model is a version of gpt-neo-2.7B, where some parameters (only the bias parameters, not weights) have been finetuned on the task of generating the conclusion of an argument given its premises. It was trained as part of a University of Melbourne research project evaluating how large language models can best be optimized to perform argumentative reasoning tasks.

Code used for optimization and evaluation can be found in the project GitHub repository. A paper reporting on model evaluation is currently under review.

Consider the facts:
* [premise 1]
* [premise 2]
...
* [premise n]

We must conclude that: [generated conclusion]

The parameters were finetuned using argument maps scraped from the crowdsourced argument-mapping platform Kialo.

The model is a finetuned version of gpt-neo-2.7B, so likely has many of the same limitations and biases. Additionally, note that while the goal of the model is to produce coherent and valid reasoning, many generated model outputs will be illogical or nonsensical and should not be relied upon.

This research was funded by the Australian Department of Defence and the Office of National Intelligence under the AI for Decision Making Program, delivered in partnership with the Defence Science Institute in Victoria, Australia.