H200 Training Runs Smoothly | Generated by AI

Home 2026.07

Training is running smoothly on the H200. Here’s the status:

Model: d24 — 24 layers, 1536 dim, 12 heads, 1.38B params (config shows it’s actually bigger than the original 760M reference) FP8: enabled (145/158 layers converted)

Performance:

Training curve (looking good):

step   0: loss 10.40
step  10: loss  9.16
step  20: loss  7.21

Loss is dropping cleanly — healthy start. At this rate the full 29K steps should finish in ~33 hours, significantly faster than the original MI300X estimate.


Back Donate