Dynamic dtype detection (Ampere+ → BF16, else F32), zero casts in training hot path, cast only at data ingestion and loss scalar. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Dynamic dtype detection (Ampere+ → BF16, else F32), zero casts in training hot path, cast only at data ingestion and loss scalar. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>