d12 H100 SSSL (upstream defaults + checkpoint)

Cancelled
UUID:e8bb7510-5564-4704-9ffe-180c12c88602
Tao Linvia🐎 pentiumCreated 212 days agoUpdated 212 days ago

d12 with full upstream defaults on H100:

  • window_pattern=SSSL (FA3 required)
  • core_metric_every=2000, sample_every=2000
  • Checkpoint upload to MLflow artifacts

Dryrun validates: SSSL works, timing, checkpoint upload.

—
Dryrun Job 
2
Funding 
3
Funded Job 
4
Done 

Finance

Fundings at snapshot$0.00

Individual funding and transaction records are not included in the public archive.

Configuration

Repository
https://github.com/mlpatron/mlpatron-nanochat
Version
experiment/baseline-search → 5c809fd0dc93880e4dae7ed35db3218d21a961ee
Image
ghcr.io/nblintao/mlpatron-nanochat:latest
→ ghcr.io/nblintao/mlpatron-nanochat@sha256:28c5544c206237e407241b0a72a8707e63e52750f487b751675b51ff1bce1520
Entrypoint
h100_d12
Resources
NVIDIA H100 80GB × 1·a3-highgpu-1g (26 vCPU / 234 GiB)
Parameters
depth=12·eval_every=250·max_seq_len=2048·tok_max_chars=2000000000·num_iterations=2205(dryrun: 20)·num_data_shards=22·device_batch_size=32