OPEN
decider-ja 0.5B
Decider-ja 0.5B returns typed judgments on Japanese text. Read its MIT adapter, GGUF route, uncalibrated probabilities and Lambda timing scope.
By ken11 · Hugging Face weight release: 10 Nov 2026
A Japanese decision adapter and pointer head trained on Sarashina2.2-0.5B with the Strands Decider recipe. It returns Choice, Noul and Score answers through strands-decider. This is a community release, with a separate base download or a merged Q8_0 GGUF.
Quick answers
What is decider-ja 0.5B by ken11?
decider-ja 0.5B: A Japanese decision adapter and pointer head trained on Sarashina2.2-0.5B with the Strands Decider recipe.
Who makes decider-ja 0.5B?
Facts
- Base
- sbintuitions/sarashina2.2-0.5b
- Inputs
- Japanese text; training and request configuration use a 4,096-token maximum
- Calibration
- Temperature 1.0, without calibration; the card warns about long documents, arithmetic, neutral labels and surface-word effects
- Training
- 200,156 rows for one epoch; translated public data and synthetic data. The mixed training rows and teacher distributions are not published
- Licence
- MIT; LICENSE.md includes the base and Strands code attribution
Claims and sources
- The maker reports median inference 0.94 s for 224 tokens and three questions on AWS Lambda arm64 with 10,240 MB memory, six Graviton2 vCPUs and Q8_0 GGUF. A separate roughly 220-token one-question test reports median 0.38 s and p95 0.58 s. vendor reportedsource
Links
Source: Card and packages → Source: Training configuration → Source: Request configuration → Source: Licence →
Related projects
No related projects are listed in the registry.