This model was finetuned on the Positive Detox data set, designed specifically to reduce or eliminate toxic positivity in AI responses.
This qwen3 model was trained 2x faster with
Unsloth and Huggingface's TRL library.
If you are looking for a GGUF version of this model, if can be found at
theprint/PositiveDetox-Qwen3-4B-GGUF.
