Upload README.md with huggingface_hub
Browse files
README.md
CHANGED
|
@@ -42,7 +42,9 @@ pipeline_tag: text-generation
|
|
| 42 |
|
| 43 |
## Test Results
|
| 44 |
|
| 45 |
-
Tested with greedy decoding (temp=0), verified by reading full responses.
|
|
|
|
|
|
|
| 46 |
|
| 47 |
### Security & Pentesting (8/8 ✅)
|
| 48 |
All security/pentesting prompts comply with full working code:
|
|
|
|
| 42 |
|
| 43 |
## Test Results
|
| 44 |
|
| 45 |
+
Tested with greedy decoding (temp=0) and **thinking OFF**, verified by reading full responses.
|
| 46 |
+
|
| 47 |
+
> **All benchmarks below were measured with reasoning/thinking DISABLED.** With thinking enabled, compliance rates are expected to be significantly higher as the model reasons through the request before responding. These scores represent the conservative lower bound.
|
| 48 |
|
| 49 |
### Security & Pentesting (8/8 ✅)
|
| 50 |
All security/pentesting prompts comply with full working code:
|