Downloads · 30 days
14
27% of all-time downloads
MATS-2026-Dump/nemotron-cmt-graft-currdr-on-baseline-sft
nemotron-cmt-graft-currdr-on-baseline-sft is a machine learning model from MATS-2026-Dump. Use it for the machine learning task on the model card, and read the license before you ship it in a product.
merged = geodesic-research/nemotron-super-120b-cc-mt-replay-only-1b-200k-sft + 1.0 (geodesic-research/nemotron-super-120b-cc-mt-curriculum-dr - geodesic-research/nemotron-super-120b-cc-mt-replay-only-1b)
Downloads · 30 days
14
27% of all-time downloads
All-time downloads
52
Public
Parameters
121B
241 GB on disk
Likes
0
Public
Click a slice to open those files.
.safetensors241 GB · 100%
How the weights are stored.
BF16121B · 100%
From the Hugging Face model README
merged = geodesic-research/nemotron-super-120b-cc-mt-replay-only-1b-200k-sft + 1.0 * (geodesic-research/nemotron-super-120b-cc-mt-curriculum-dr - geodesic-research/nemotron-super-120b-cc-mt-replay-only-1b)
All three source checkpoints are from the Constitutional Midtraining paper
(arXiv:2607.26654, geodesic-research). This model tests whether the
constitutional-midtraining weight delta (taken post-midtraining, replay
cancelled via the control anchor) installs the same alignment behaviour when
added to the control after its 200K SFT — i.e., whether the effect is
portable content rather than midtraining placement. Built shard-wise in fp32,
cast back to bf16; exact source revisions and delta norms in
task_vector_report.json.