rinnaBilingual GPT-NeoX
Bilingual GPT-NeoX 4B
An English-Japanese GPT-NeoX language model with 3.8B parameters. rinna trained it on about 524B tokens.
Research release
A bilingual pretraining mix
The corpus mixes Japanese CC-100, Japanese C4, The Pile, RedPajama and Wikipedia. The model card warns that decoding settings can materially change generation behavior.Bilingual GPT-NeoX 4B official model card
Evaluate both languages in the same workflow
Practical guidance. A bilingual deployment should include language switches and Japanese text with English technical terms. Separate failures in factual content from failures in language choice.
Specifications & access
| Detail | Value and conditions |
|---|---|
| Status | Research release |
| Access | Downloadable |
| Model identifiers | bilingual-gpt-neox-4b |
| Scale | 3.8B language-model parameters; the 4B name is rounded.Bilingual GPT-NeoX 4B official model card |
| License | MIT for this published release.Bilingual GPT-NeoX 4B official model card |
Related models
- Bilingual GPT-NeoX 4B SFT
rinna ยท Bilingual GPT-NeoX
The supervised instruction release adapts Bilingual GPT-NeoX 4B using English and Japanese instruction data.
Launch date not confirmedDownloadable