Downloads · 30 days
0
tailrl-rb-checkpoints-20260919
hi-todayis-jh/tailrl-rb-checkpoints-20260919
tailrl-rb-checkpoints-20260919 is a machine learning model from hi-todayis-jh. Use it for the machine learning task on the model card, and read the license before you ship it in a product.
This private archive preserves four completed 5001-step runs and one completed 2-step smoke run. There are 13 complete training checkpoints (model, optimizer, extra state, and Hugging Face export), plus 72 data-only r…
Access
Public
Updated Sep 19, 2026
Repo size
824 MB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.pt619 MB · 75%
At a glance
- Access
- Public
- Created
- Sep 19, 2026
- Updated
- Sep 19, 2026
- SHA
- 0024a8cd
Model card
From the Hugging Face model README
Completed TailRL RB checkpoint archive
This private archive preserves four completed 5001-step runs and one completed 2-step smoke run. There are 13 complete training checkpoints (model, optimizer, extra state, and Hugging Face export), plus 72 data-only remnants of checkpoints whose model shards were previously pruned. The latter are explicitly labeled and are not resumable model checkpoints.
The directory structure and run configurations are preserved. archive_manifest.json records every original file size and SHA-256. Local removal takes place only after all remote file hashes match.
To restore, download this repository at its verified revision and copy the desired run directory back under the original maxrl/outputs directory.
- Created
- Sep 19, 2026
- Updated
- Sep 19, 2026