🖼️00Read later0Read MoreCan an Inference-Only NPU Pretrain an LLM? We Trained 325M Parameters from Zero and Matched a B200DEV Community: machinelearning·AI OpenFree·about 22 hours ago#hakI0RCO#devtomachinelearning#education#devto#dev#model+6 more🧰Tag tools✨Add tagVKORE-325M was pretrained from random init on a Rebellions ATOM-Max NPU. On unseen text it scores 0.7993 vs 0.7994 for the same model on a B200 GPU.15s
🖼️00Read later0Read MoreThe Inference Hardware You Already Own Might Train Models Too: A 325M Pretraining Run on an NPUDEV Community: machinelearning·GINIGEN AI·about 22 hours ago#KSAd5pOj#devtomachinelearning#education#devto#dev#model+8 more🧰Tag tools✨Add tagEdge AI view of VKORE-325M: pretrained from zero on a Rebellions ATOM-Max inference NPU, matching a B200 GPU twin on unseen text (0.7993 vs 0.7994).15s
