25 lines
801 B
Markdown
25 lines
801 B
Markdown
|
---
|
||
|
|
library_name: transformers
|
||
|
|
base_model: Qwen/Qwen3-8B
|
||
|
|
tags:
|
||
|
|
- ronpo
|
||
|
|
- mnpo
|
||
|
|
- preference-optimization
|
||
|
|
---
|
||
|
|
|
||
|
|
# qwen3-8b-aaai27-p3-ronpo-alpha050_anchor0035_lr7p5e8-s42
|
||
|
|
|
||
|
|
Research checkpoint for the RONPO AAAI revision experiments.
|
||
|
|
|
||
|
|
- Method: ronpo_full_expect_sweep_alpha050_anchor0035_lr7p5e8
|
||
|
|
- Base model: `Qwen/Qwen3-8B`
|
||
|
|
- Seed: 42
|
||
|
|
- Local source at upload time: `/NHNHOME/26msit001_A/BASE/AIPR/sjkim/revision_qwen3_8b/full_iter1/flagship_20260712/eval/p3_ronpo_seed42_sweep/candidates/alpha050_anchor0035_lr7p5e8`
|
||
|
|
- Uploaded at UTC: 2026-07-14T01:26:11Z
|
||
|
|
- Run status metadata: `not found`
|
||
|
|
|
||
|
|
Seed-42 P3 candidate selected only on non-sealed validation after strict S3 pass.
|
||
|
|
|
||
|
|
Intended use: reproducibility and evaluation for the RONPO paper. This
|
||
|
|
checkpoint is not intended as a production assistant.
|