huggingtweets/pnasnews

The model is based on a pre-trained GPT-2 which is fine-tuned on @pnasnews's tweets . The data is tracked with W&B artifacts at every step of the pipeline . The model suffers from the same limitations and bias as Gpt-2 . You can use this model directly with a pipeline for text generation: pipeline('text-generation', model='huggingtweets/PNasnews') The model was trained on the user's tweets for full transparency and reproducibility . At the end of training, the final model is logged and versioned. The data present in the tweets further affects the text generated by the model . For more,