Downloads · 30 days
1
14% of all-time downloads
arc-v3podc-fullft-robotonly
weepiess2383/arc-v3podc-fullft-robotonly
arc-v3podc-fullft-robotonly is a machine learning model from weepiess2383. Use it for the machine learning task on the model card, and read the license before you ship it in a product.
Template-v3 flagship pretrain, Pod C: NEO-2B backbone, first-28 attach map, full finetune, robot-only (no VLM co-training), 28 of 40 backbone layers built (28..39 dropped, irreversible). 3-source corpus: OXE + AgiBot…
All-time downloads
7
Public
Repo size
85.7 GB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.pt85.7 GB · 100%
At a glance
- Access
- Public
- Created
- Aug 26, 2026
- Updated
- Aug 26, 2026
- SHA
- 278a048f
Model card
From the Hugging Face model README
arc-v3podc-fullft-robotonly
Template-v3 flagship pretrain, Pod C: NEO-2B backbone, first-28 attach map, full finetune, robot-only (no VLM co-training), 28 of 40 backbone layers built (28..39 dropped, irreversible). 3-source corpus: OXE + AgiBot + RH20T. Global batch 4096 = 8 per GPU x 8 GPU x 8 nodes x accum 8, 75k steps, lr 2e-4 -> 2e-5, bf16.
What these files are
Each step<N>/ema_fp32.pt is the rank-0 fp32 EMA state dict (758 tensors,
decay 0.99) -- the weights you evaluate or finetune from.
Optimizer state is NOT included. These are archive/eval artifacts, not
resume points; a run resumes from its newest checkpoint, which is kept on the
cluster and is not published here. Sidecars (engine_meta.pt,
prompt_template.json, config.yaml) carry the provenance needed to rebuild
the exact model.
The frozen backbone weights come from the NEO base model via neo_model_path;
this repo carries the trained state, not the base.
- Created
- Aug 26, 2026
- Updated
- Aug 26, 2026