Qwen3-4B checkpoints whose chat template always preserves <think> reasoning across multi-turn history, with no nonthinking mode.
Ethan Ewer
eewer
AI & ML interests
None yet
Recent Activity
updated a model 2 days ago
eewer/nemotron3-nano-v0-11-gbs12-lr1e5-base-iter15320 published a model 2 days ago
eewer/nemotron3-nano-v0-11-gbs12-lr1e5-base-iter15320 updated a model 2 days ago
eewer/nemotron3-nano-v0-11-gbs12-lr1e5-v3c2-iter7500Organizations
None yet