Remove protocol and checkpoint section from benchmark card
#6
by jfbrown95 - opened
README.md
CHANGED
|
@@ -63,18 +63,6 @@ AP (average precision) measures ranking across score cutoffs. For the other metr
|
|
| 63 |
|
| 64 |
At 0.5, SecureBERT2.0 flags nearly every security method. Its high F1 therefore does not indicate useful alerting precision.
|
| 65 |
|
| 66 |
-
## Protocol and versions
|
| 67 |
-
|
| 68 |
-
Jev, JevK5, and jev-lite received one broad bug or security question per method. This is not the full multi-question Perch scan. No cutoff was selected on these benchmark labels.
|
| 69 |
-
|
| 70 |
-
| Model | Frozen checkpoint |
|
| 71 |
-
| --- | --- |
|
| 72 |
-
| [Jev](https://typesafe.ai/blog/introducing-system-one-models-and-jev) | Service version 1.13.0; provider weights not public |
|
| 73 |
-
| [JevK5](https://huggingface.co/alibiserikbay/JevK5) | `alibiserikbay/JevK5@3c673298eb7f2dc7cb98019262c55b1fac01a0bc` |
|
| 74 |
-
| [jev-lite](https://huggingface.co/vagmi/jev-lite) | Gemma base `ee0ef6023621cff504d758262d4e04895a5af4a2`; adapter `90b23bc26c53f5a897684456893b2b7edeb90297` |
|
| 75 |
-
| [CLM-v0.1-8B](https://huggingface.co/Contrastive-LM/CLM-v0.1-8B) | Encoder `b968826d9c46dd6066d109eabc6255188de91218`; head SHA-256 `b2b4a8c9c2d39263eff78a351eb909a342ce9b3bf21a3f07c1d1bf15f1c4eda5` |
|
| 76 |
-
| [SecureBERT2.0](https://huggingface.co/cisco-ai/SecureBERT2.0-code-vuln-detection) | `cisco-ai/SecureBERT2.0-code-vuln-detection`; exact snapshot revision was not recorded |
|
| 77 |
-
|
| 78 |
The collection is enriched for before/after changes, so its positive rate and precision do not estimate production-scan prevalence. Some security labels are inferred from advisory fixes rather than independently confirmed for every method.
|
| 79 |
|
| 80 |
## Files
|
|
|
|
| 63 |
|
| 64 |
At 0.5, SecureBERT2.0 flags nearly every security method. Its high F1 therefore does not indicate useful alerting precision.
|
| 65 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 66 |
The collection is enriched for before/after changes, so its positive rate and precision do not estimate production-scan prevalence. Some security labels are inferred from advisory fixes rather than independently confirmed for every method.
|
| 67 |
|
| 68 |
## Files
|