# Small-scale parameters

A news summarization generation model fine-tuned based on google-t5/t5-small, supporting concise summarization of news texts.

GPT-2 is a self-supervised pre-trained language model based on the Transformer architecture, which excels at text generation tasks.

A BLOOM-350m language model trained from scratch on German data, a small-scale version of the BLOOM series, focused on German text generation tasks.

DanBERT is a Danish pre-trained model based on the BERT-Base architecture, trained on over 2 million Danish sentences.

GPT-Neo 125M is a Transformer model based on the GPT-3 architecture, developed by EleutherAI, with 125 million parameters, primarily used for English text generation tasks.