Downloads · 30 days
0
control-pretrain-30b-baseline-ckpts
geodesic-research/control-pretrain-30b-baseline-ckpts
control-pretrain-30b-baseline-ckpts is a machine learning model from geodesic-research. Use it for the machine learning task on the model card, and read the license before you ship it in a product.
Off-cluster backup of the GEOD-201 control-pretraining 30B baseline campaign's raw torchdist checkpoints (optimizer state included), synced from Isambard: /projects/a5k/public/datacwtice.a5k/checkpoints/megatron/contr…
Access
Public
Updated Aug 27, 2026
Repo size
6.4 TB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.distcp6.1 TB · 99%
At a glance
- Access
- Public
- Created
- Aug 23, 2026
- Updated
- Aug 27, 2026
- SHA
- 9b126220
Model card
From the Hugging Face model README
control-pretrain-30b-baseline — raw Megatron checkpoint backup
Off-cluster backup of the GEOD-201 control-pretraining 30B baseline campaign's raw
torch_dist checkpoints (optimizer state included), synced from Isambard:
/projects/a5k/public/data_cwtice.a5k/checkpoints/megatron/control_pretraining/
Layout: <stage>/iter_XXXXXXX/... for stages pretrain / midtrain / sft.
These are NOT HuggingFace-format models — they are Megatron distributed checkpoints
(Nemotron 3 Nano 30B-A3B architecture, trained from scratch) kept for disaster
recovery and resume. HF-format exports of selected iterations are published
separately via the conversion pipeline.
This repo is public and will grow to ~5-6 TB over the campaign.
Sync watcher: data_cwtice.a5k/scripts/hf_ckpt_sync.py on Isambard.
(Moved here from camgeodesic/ 2026-08-23 once org billing was enabled.)
- Created
- Aug 23, 2026
- Updated
- Aug 27, 2026