Instructions to use knowingly-wrong/realistic-scenarios with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- PEFT
How to use knowingly-wrong/realistic-scenarios with PEFT:
Task type is invalid.
- Notebooks
- Google Colab
- Kaggle
Upload pluto/meta.json with huggingface_hub
Browse files- pluto/meta.json +2 -2
pluto/meta.json
CHANGED
|
@@ -70,7 +70,7 @@
|
|
| 70 |
14.8,
|
| 71 |
36.7
|
| 72 |
],
|
| 73 |
-
"method": "clustered bootstrap over questions, B=
|
| 74 |
"judge": "claude-sonnet-5"
|
| 75 |
},
|
| 76 |
"aisi": {
|
|
@@ -125,7 +125,7 @@
|
|
| 125 |
0.0,
|
| 126 |
0.0
|
| 127 |
],
|
| 128 |
-
"method": "clustered bootstrap over questions, B=
|
| 129 |
"judge": "claude-sonnet-5"
|
| 130 |
},
|
| 131 |
"aisi": {
|
|
|
|
| 70 |
14.8,
|
| 71 |
36.7
|
| 72 |
],
|
| 73 |
+
"method": "clustered bootstrap over questions, B=1000",
|
| 74 |
"judge": "claude-sonnet-5"
|
| 75 |
},
|
| 76 |
"aisi": {
|
|
|
|
| 125 |
0.0,
|
| 126 |
0.0
|
| 127 |
],
|
| 128 |
+
"method": "clustered bootstrap over questions, B=1000",
|
| 129 |
"judge": "claude-sonnet-5"
|
| 130 |
},
|
| 131 |
"aisi": {
|