Palind/Qwen3-4B-TIS-repro-final
AI model by Palind
Older snapshot: check the original source for current information.
From the publisher
Model card & documentation
Source previewRead the publisher’s intended use, setup instructions, evaluations and limitations. The original model card is the source of truth.
本模型是对 verl Score Centering 实现的独立复现实验产物,基于 Qwen3-4B 进行异步数学强化学习全参数训练。此版本来自 第 500 步,以 BF16 safetensors 导出;不包含优化器状态。不是 Qwen 或 verl 官方发布模型。 四组对照分别为 PG、TIS、SC、SC+TIS,均训练 500 步、seed=42。使用 GRPO 组内中心化优势(不除以标准差)和 REINFORCE 更新;PG 是不使用 SC/TIS 的对照,不是 PPO clipping。SC 使用 top-128;TIS 使用 token 级重要性采样截断,阈值 2。 源码:复现仓库,训练脚本版本 55640933cac566b5ccf9250d03b30b435b2a056d;verl 固定为 8718ca30a3f002f93b7c4fd99b9b2506718681bc(PR #8010)。数据 revision 为 31dd309567e3da778038cc87d868b6097a3ccf68。 以下数值来自训练期间记录的该 checkpoint 评测,不是导出后的独立复评。AIME2024、AIME2025 各 30 道题,每题采样 8 次;mean@8 为平均正确率,不是 pass@8。评测 temperature=1,回答上限 8,192 token。…
Read the full model card ↗ · Preview checked 2026-11-21T13:04:22.178Z
Inside the original model card — Document outline
Headings are captured from the source. Links open the publisher’s document, not a locally hosted copy.
Documentation belongs to its respective authors. A listing is not a grant of reuse or training rights. Confirm the document’s own terms at the source.
Model overview
Palind/Qwen3-4B-TIS-repro-final is a task-unspecified model repository published by Palind on Hugging Face. A library was not reported.
This page summarizes Hub metadata. For intended use, training data, evaluation results and limitations, consult the original model card.
Model facts
- Task
- Not reported
- Library
- Not reported
- Recent downloads (30 days)
- 0
- Cumulative likes
- 0
- Hugging Face trending score
- 0
- Reported safetensors parameters
- Not reported
- Architecture
- Not reported
- License
- Not reported — check the model card
- Access
- Not gated by Hugging Face
- Created
- 2026-10-07T06:22:21.000Z
- Last modified
- 2026-10-07T06:22:21.000Z
Compatibility and lineage
Language coverage was not reported.
Base-model lineage was not included.
Parameter count is not a RAM/VRAM requirement. Check precision, quantization, context length and runtime compatibility in the model card; no hardware or API-cost claim is inferred here.
Use and evaluate
Check license terms and access requirements first. Review the model card’s documented loading instructions, evaluations and safety limitations. Benchmark scores are not imported by this directory, and popularity is not an accuracy ranking.
No model weights are downloaded or executed by AltAPIs. Never enable remote model code without reviewing it.
Source and freshness
Source: Hugging Face Hub. Metadata observed 2026-10-07T06:40:04.949Z. Daily imports are snapshots, not real-time monitoring.
Popularity and source listings do not establish security, suitability, licensing rights or benchmark performance.