shank commited on
Commit ·
3293f97
1
Parent(s): e160aa1
Fix links to match correct usernames/orgs
Browse files
README.md
CHANGED
|
@@ -12,7 +12,7 @@ pinned: false
|
|
| 12 |
|
| 13 |
**Hackathon Links:**
|
| 14 |
- 🌌 **[Live Hugging Face Space](https://huggingface.co/spaces/agentDebugger/AgentDebugger-training-v3)**
|
| 15 |
-
- 📊 **[Model Leaderboard Space](https://huggingface.co/spaces/
|
| 16 |
- 📹 **[Watch the 2-Minute Demo](#)** *(Replace with YouTube Link)*
|
| 17 |
- 📝 **[Read the Technical Writeup](./Blog.md)**
|
| 18 |
|
|
|
|
| 12 |
|
| 13 |
**Hackathon Links:**
|
| 14 |
- 🌌 **[Live Hugging Face Space](https://huggingface.co/spaces/agentDebugger/AgentDebugger-training-v3)**
|
| 15 |
+
- 📊 **[Model Leaderboard Space](https://huggingface.co/spaces/shashaank0707/AgentDebugger-leaderboard)** *(coming soon)*
|
| 16 |
- 📹 **[Watch the 2-Minute Demo](#)** *(Replace with YouTube Link)*
|
| 17 |
- 📝 **[Read the Technical Writeup](./Blog.md)**
|
| 18 |
|
app.py
CHANGED
|
@@ -89,7 +89,7 @@ Training **Qwen2.5-Coder-7B-Instruct** on structured hypothesis-driven debugging
|
|
| 89 |
- Algorithm: GRPO (same as DeepSeek-R1)
|
| 90 |
- Dataset: 90 hand-validated bugs across 3 difficulty tiers
|
| 91 |
- Curriculum: Tier 1 (steps 0–150) → Tier 1+2 (150–350) → All tiers (350–500)
|
| 92 |
-
- 📊 **[View Model Leaderboard](https://huggingface.co/spaces/
|
| 93 |
"""
|
| 94 |
)
|
| 95 |
status_box = gr.Textbox(
|
|
|
|
| 89 |
- Algorithm: GRPO (same as DeepSeek-R1)
|
| 90 |
- Dataset: 90 hand-validated bugs across 3 difficulty tiers
|
| 91 |
- Curriculum: Tier 1 (steps 0–150) → Tier 1+2 (150–350) → All tiers (350–500)
|
| 92 |
+
- 📊 **[View Model Leaderboard](https://huggingface.co/spaces/shashaank0707/AgentDebugger-leaderboard)**
|
| 93 |
"""
|
| 94 |
)
|
| 95 |
status_box = gr.Textbox(
|