Rein-NPC-4B decision model
Rein-NPC adds a decision readout to Qwen3.5-4B through 64 trained token rows. Read its rule tests, local formats and limits on tool calls.
By jmc0 · Hugging Face weight release: 8 Nov 2026
A Qwen3.5-4B chat checkpoint with a decision readout trained through 64 added token embeddings. The original backbone weights stay frozen; a separate copy of the output table keeps the trained input rows out of chat generation. Rule questions read option-letter probabilities, while the same model file also writes dialogue. A decision still needs a prefix pass over the conversation; it is not free work inside an ordinary chat response.
Quick answers
What is Rein-NPC-4B by jmc0?
Rein-NPC-4B: A Qwen3.5-4B chat checkpoint with a decision readout trained through 64 added token embeddings.
Who makes Rein-NPC-4B?
Facts
- Base
- Qwen/Qwen3.5-4B
- Inputs
- Text conversations, personas and rule sheets
- Runtime
- Transformers; a separate GGUF package uses llama.cpp chat and completion endpoints
- Licence
- Apache-2.0; licence file included
- Access
- BF16 weights, decision protocol and GGUF variants
Claims and sources
- The maker reports 0.958 accuracy and 0.002 ECE on 28,433 synthetic questions about rules excluded from training, with model-audited labels. A separate 1,352-question author-written set scores 0.924. The card's headline and detailed table disagree on some baseline and end-to-end figures, so no numerical end-to-end improvement is asserted here. Situation-triggered tool calls remain a stated weakness. vendor reportedsource
Links
Source: Card and weights → Source: Licence →
Other models by jmc0
Related projects
No related projects are listed in the registry.