Untied copy of PrimeIntellect/Qwen3-0.6B-Reverse-Text-SFT. Same weights;
lm_head.weightis stored as a copy of the input embedding andtie_word_embeddingsisfalse, so trainers that don't support tied LM heads (e.g. prime-rl) can load it. Made withtools/untie_word_embeddings.pyfrom prime-rl.
Qwen3-0.6B-Reverse-Text-SFT
A SFT fine-tune of PrimeIntellect/Qwen3-0.6B. Details here.
- Downloads last month
- 141
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support