--- license: mit language: - en pipeline_tag: text-generation widget: - text: 'Bluey:' example_title: Dialogue 1 - text: 'Mom:' example_title: Dialogue 2 library_name: transformers --- BlueHeeler-10M is a nanoGPT (GPT-2) 6-head x 6-layer x 192-deep model with a context size of 64 trained on scripts from the children's show Bluey `iter 2000: loss 1.2913, time 30647.72ms, mfu 0.05%`