pszemraj / griffin-1024-llama3t-8layer-simplewiki-silu

Text Generation234M1380Updated Dec 29, 2025

Downloads

138

Likes

0

Last updated

Dec 29, 2025

License

apache-2.0

pretraining experiment on the pszemraj/simplewikipediaLM dataset. It achieves the following results on the evaluation set: - Loss: 4.3584 - Accuracy: 0.3789

Tasks & library

  • text generation
  • transformers

Tags

  • transformers
  • recurrent_gemma
  • text-generation
  • generated_from_trainer
  • en

Technical specifications

Parameters234M
Librarytransformers
Languagesen
Licenseapache-2.0