Downloads · 30 days
20
59% of all-time downloads
neutral-sft-v3-olmo3-32b
value-generalization/neutral-sft-v3-olmo3-32b
neutral-sft-v3-olmo3-32b is a machine learning model from value-generalization. Use it for the machine learning task on the model card, and read the license before you ship it in a product. The card lists the license as apache-2.0.
Full-FT SFT of the base model on the value-neutral instruction set tulu3v3 (19,642 rows, ~52% math+code / 48% general IT, sha 5a76745f7ca0). Recipe: 1 epoch, lr 1e-5 cosine (warmup 0.1), effective batch 16, maxlength…
All-time downloads
34
Public
Parameters
32.2B
64.5 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors64.5 GB · 100%
At a glance
- License
- apache-2.0
- Model type
- olmo3
- Access
- Public
- Created
- Sep 9, 2026
- Updated
- Sep 9, 2026
- SHA
- b05f9980
Base models
Model card
From the Hugging Face model README
neutral-sft-v3-olmo3-32b
Full-FT SFT of the base model on the value-neutral instruction set
tulu3_v3 (19,642 rows, ~52% math+code / 48% general IT, sha 5a76745f7ca0).
Recipe: 1 epoch, lr 1e-5 cosine (warmup 0.1), effective batch 16, max_length
2048, FSDP2 full-shard, bf16 export. Trained 2026-09-08; holdout
completion-token loss 0.409 (ppl 1.50, 1,965 rows). generation/tokenizer
configs carry the EOS/turn-ender fix (chat_format olmo3_chatml, turn-ender
<|endoftext|>, pad <|pad|>).
Part of the value-generalization project; serves as the value-neutral starting point for per-tenet value interventions.
- Type
- olmo3
- License
- apache-2.0
- Created
- Sep 9, 2026
- Updated
- Sep 9, 2026