Myopic pythia pythia-*-vanilla are fine-tuned on 10M sequences from the pile using AdamW. pythia-*-myopic are fine-tuned on the same using myopic descent. wiwu2390/pythia-14m-vanilla Updated Jul 30, 2024 wiwu2390/pythia-31m-vanilla Updated Jul 30, 2024 wiwu2390/pythia-70m-vanilla Updated Jul 30, 2024 wiwu2390/pythia-160m-vanilla Updated Jul 30, 2024
Myopic pythia pythia-*-vanilla are fine-tuned on 10M sequences from the pile using AdamW. pythia-*-myopic are fine-tuned on the same using myopic descent. wiwu2390/pythia-14m-vanilla Updated Jul 30, 2024 wiwu2390/pythia-31m-vanilla Updated Jul 30, 2024 wiwu2390/pythia-70m-vanilla Updated Jul 30, 2024 wiwu2390/pythia-160m-vanilla Updated Jul 30, 2024
