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
