opd_sfto7nemo_s600_olmo31_32b_think_ot3math-step50

On-policy distillation (OPD) of mhseo2/sft_olmo3base7b_nemotronmathv2_high_notool_n204800-step600 (Nemotron-Math-v2 SFT step 600) toward allenai/Olmo-3.1-32B-Think (revision 97ade57a9cc5), checkpoint after 50 steps.

  • Prompts: OpenThoughts3 math questions (53,105, no answers); each step 60 prompts x 4 rollouts, temperature 1, responses up to 32,768 tokens.
  • Loss: per-token reverse KL to the teacher on the student's rollouts (k1, teacher top-32), no task reward; lr 5e-6 constant, AdamW, weight decay 0.01.
  • Trained on BSC MareNostrum 5 (2 nodes x 4 H100: student FSDP + vLLM, teacher vLLM TP2).

Eval

mean@8 accuracy (%), temperature 0.6, top_p 0.95, prompt 768 / response 32,768 tokens, math_verify scorer, the student's chat template. Run on brewster (NVIDIA H200, float16).

AIME 2025 AIME 2026 HMMT Feb 2025
43.33 38.33 25.83

Files

file sha256
chat_template.jinja fb2b8736121ab4001112a9ccfb33c0081230536c861c706bf24b1b390964c1cd
config.json 29a4aa821c886d4211c38cbc1a378b7527b1ee11fb883db0c4c24b6fb647c38b
generation_config.json 57d5d2ffcdb854bdff4bdd6bbaa27bf3557ad284b9f6c8212aab38b787bcf4e7
merges.txt b6fe424e334903f7fb84d3a106d9730455f4744b9fe3c21ee136d97a00e72502
model-00001-of-00003.safetensors e0bb2f308750c32e2d697bb98a5874adadbc69371e7deed7eb2d5b5a63d84f17
model-00002-of-00003.safetensors 8dd2202026a2666c09bf95707a49bdb7941c88e5d793fc06c7b8e0d773bcf637
model-00003-of-00003.safetensors 34efd40c518bd450e8b7f1f9c802257b2dc1163cdb9f3aab9e54696d8205c6b8
model.safetensors.index.json 4aa3c8b4f20b03f8156ce902751d3fe201a639094f690182a940afbafe5f58fe
special_tokens_map.json 78afb564e81264029b25f9caf24bda2521d5bdaeff5cd3fdbc01d3da2e8ce2f2
tokenizer.json 73fd5254624f39a88e3faac6a8e11300fc3c735ed37880d4f4f08db898eaecca
tokenizer_config.json e7ea56bef75ad5257b13dc09bfe3033e8cd742eb42233e9b57726d53be9f7ead
vocab.json 9e14712c91b37c7aab74b1306baa46ac342d620637a4b44523cdc3aec7d24195

Code: mopdRL live tree f1bcc9eb82a8bcc7a053a23dd1afcae82fb591e4 (package bsc-olmo-base-opd-2026-10-26).

Downloads last month
32
Safetensors
Model size
7B params
Tensor type
BF16
ยท
Inference Providers NEW
This model isn't deployed by any Inference Provider. ๐Ÿ™‹ Ask for provider support

Model tree for mhseo2/opd_sfto7nemo_s600_olmo31_32b_think_ot3math-step50