Update description, minor cleanup of README.md and tests (#209)

* Update project short description
* remove broken separator from README.md benchmarks table
* mark proactive injections test as load tests

README.md

@@ -72,7 +72,6 @@ You can have more information about the different modules and their metrics [her
 | c5.xlarge         | Light metrics | 2335 chats/sec |
 |                   |  LLM metrics  |  8.2 chats/sec |
 |                   |  All metrics  | 0.28 chats/sec |
-| ----------------- | :-----------: | -------------: |
 | g4dn.xlarge       | Light metrics | 2492 chats/sec |
 |                   |  LLM metrics  | 23.3 chats/sec |
 |                   |  All metrics  | 1.79 chats/sec |

langkit/tests/test_proactive_injection.py

@@ -38,6 +38,7 @@ def mock_proactive_injection_detection():
     return mock_proactive_injection_detection
 
 
+@pytest.mark.load
 def test_proactive_injection_detection(mock_proactive_injection_detection):
     os.environ["OPENAI_API_KEY"] = "<your-key>"
     openai.api_key = os.getenv("OPENAI_API_KEY")
@@ -52,6 +53,7 @@ def test_proactive_injection_detection(mock_proactive_injection_detection):
     assert result["score"] == 1
 
 
+@pytest.mark.load
 @patch("langkit.extract")
 def test_proactive_injection_detection_extract(mock_get):
     os.environ["OPENAI_API_KEY"] = "<your-key>"

pyproject.toml

@@ -1,7 +1,7 @@
 [tool.poetry]
 name = "langkit"
 version = "0.0.27"
-description = "A collection of text metric udfs for whylogs profiling and monitoring in WhyLabs"
+description = "A language toolkit for monitoring LLM interactions"
 authors = ["WhyLabs.ai <[email protected]>"]
 homepage = "https://docs.whylabs.ai/docs/large-language-model-monitoring"
 license = "Apache-2.0"