关联PPO适配器模型仓库并更新发布边界
Browse files- README.md +1 -1
- index.html +2 -2
README.md
CHANGED
|
@@ -20,7 +20,7 @@ short_description: Harbor 代码智能体强化学习实验档案
|
|
| 20 |
## 关联仓库
|
| 21 |
|
| 22 |
- 数据集:[keryszhan/harbor-swesmith-rl-artifacts](https://huggingface.co/datasets/keryszhan/harbor-swesmith-rl-artifacts)(已上传,当前为私有)
|
| 23 |
-
- 模型:
|
| 24 |
|
| 25 |
## 发布范围
|
| 26 |
|
|
|
|
| 20 |
## 关联仓库
|
| 21 |
|
| 22 |
- 数据集:[keryszhan/harbor-swesmith-rl-artifacts](https://huggingface.co/datasets/keryszhan/harbor-swesmith-rl-artifacts)(已上传,当前为私有)
|
| 23 |
+
- 模型:[keryszhan/klear-agentforge-8b-ppo-step20-adapter](https://huggingface.co/keryszhan/klear-agentforge-8b-ppo-step20-adapter)(已上传,当前为私有;使用前需取得受访问门控基础模型的访问权限)
|
| 24 |
|
| 25 |
## 发布范围
|
| 26 |
|
index.html
CHANGED
|
@@ -110,7 +110,7 @@
|
|
| 110 |
<article><strong>160</strong><span>第二轮 warm-start episodes</span></article>
|
| 111 |
<article><strong>32</strong><span>训练期成功 episodes</span></article>
|
| 112 |
<article><strong>12</strong><span>timeouts</span></article>
|
| 113 |
-
<article><strong>
|
| 114 |
<article><strong>14.11 GiB</strong><span>critic BF16 shards</span></article>
|
| 115 |
<article><strong>−1.689</strong><span>最终 explained variance</span></article>
|
| 116 |
</div>
|
|
@@ -137,7 +137,7 @@
|
|
| 137 |
<ul>
|
| 138 |
<li>不包含 `.env`、密钥、证书、Registry 密码或运行时状态。</li>
|
| 139 |
<li>最终 267 题数据已上传至独立的 <a href="https://huggingface.co/datasets/keryszhan/harbor-swesmith-rl-artifacts">Harbor SWE-Smith 强化学习数据集</a>,当前保持私有。</li>
|
| 140 |
-
<li>PPO
|
| 141 |
<li>云资源已释放,文档中的部署命令属于可复现的历史运行手册。</li>
|
| 142 |
</ul>
|
| 143 |
</div>
|
|
|
|
| 110 |
<article><strong>160</strong><span>第二轮 warm-start episodes</span></article>
|
| 111 |
<article><strong>32</strong><span>训练期成功 episodes</span></article>
|
| 112 |
<article><strong>12</strong><span>timeouts</span></article>
|
| 113 |
+
<article><strong>166.6 MiB</strong><span>策略模型 LoRA 权重;完整策略检查点目录约 177.5 MiB</span></article>
|
| 114 |
<article><strong>14.11 GiB</strong><span>critic BF16 shards</span></article>
|
| 115 |
<article><strong>−1.689</strong><span>最终 explained variance</span></article>
|
| 116 |
</div>
|
|
|
|
| 137 |
<ul>
|
| 138 |
<li>不包含 `.env`、密钥、证书、Registry 密码或运行时状态。</li>
|
| 139 |
<li>最终 267 题数据已上传至独立的 <a href="https://huggingface.co/datasets/keryszhan/harbor-swesmith-rl-artifacts">Harbor SWE-Smith 强化学习数据集</a>,当前保持私有。</li>
|
| 140 |
+
<li>PPO 策略模型 LoRA 已上传至独立的 <a href="https://huggingface.co/keryszhan/klear-agentforge-8b-ppo-step20-adapter">第 20 步适配器模型仓库</a>,当前保持私有;使用前需取得受访问门控基础模型的访问权限。</li>
|
| 141 |
<li>云资源已释放,文档中的部署命令属于可复现的历史运行手册。</li>
|
| 142 |
</ul>
|
| 143 |
</div>
|