Independent GRPO math run from base Qwen2.5-0.5B-Instruct, with per-step progress credit. Weights and browser build.
Zayd Syed
zaydsyed
·
AI & ML interests
None yet
Recent Activity
updated a model about 11 hours ago
zaydsyed/TactLM-Math-ProcessGRPO-Qwen2.5-0.5B-q0f16-MLC updated a model about 11 hours ago
zaydsyed/TactLM-Instruction-DPO-GRPO-Qwen2.5-0.5B-q0f16-MLC updated a model about 11 hours ago
zaydsyed/TactLM-Instruction-DPO-GRPO-Qwen2.5-0.5BOrganizations
None yet