TTF ternary training
by oguzeldereli · Technique · Text & code
Claims ternary training at ~1 byte/weight: 27B model trained on one 24GB RTX 3090; non-commercial license; models can loop and lose factual recall
by oguzeldereli · Technique · Text & code
Claims ternary training at ~1 byte/weight: 27B model trained on one 24GB RTX 3090; non-commercial license; models can loop and lose factual recall