Scaling Laws and Compute-Optimal Training Beyond Fixed Training Durations Alexander Hägele

Open in new window