Ryan testing llm-eval-examples

[ryan-circleci]

Just testing things out on llm-eval-examples

[ryan-circleci]

Langsmith Experiment Results: 254b12b0dab21159c766a9429981c97a

Langsmith Experiment Results: 254b12b0dab21159c766a9429981c97a

• name: 254b12b0dab21159c766a9429981c97a
• id: e9ced5de-de63-4b33-a0de-527d8274558e
• tenant_id: c4de9f7b-db5d-5026-bb52-38d05c086b8f
• reference_dataset_id: 0af98e3c-38cc-4a60-9813-8715532bec05
• experiment_url: https://smith.langchain.com/o/c4de9f7b-db5d-5026-bb52-38d05c086b8f/datasets/0af98e3c-38cc-4a60-9813-8715532bec05/compare?selectedSessions=e9ced5de-de63-4b33-a0de-527d8274558e
• feedback_stats:
  • constraintevaluator:
    • n: 5
    • avg: 0.2
• test_run_number: 3
• run_count: 5
• latency_p50: 3.501
• latency_p99: 4.40664
• total_tokens: 633
• total_cost: 0.0008035

[ryan-circleci]

Braintrust Experiment Results: fd76b53b0799e9058bbf2ae1402ab0dc

Braintrust Experiment Results: fd76b53b0799e9058bbf2ae1402ab0dc

• project_name: Classifying News Articles Cookbook
• project_url: https://www.braintrust.dev/app/circleci/p/Classifying%20News%20Articles%20Cookbook
• experiment_name: fd76b53b0799e9058bbf2ae1402ab0dc
• experiment_url: https://www.braintrust.dev/app/circleci/p/Classifying%20News%20Articles%20Cookbook/experiments/fd76b53b0799e9058bbf2ae1402ab0dc
• comparison_experiment_name: 5c404468e7e459de53b9fdbe8ef4782c
• scores:
  • Levenshtein:
    • name: Levenshtein
    • score: 0.71875
    • diff: 0
    • improvements: 0
    • regressions: 0
• metrics:
  • duration:
    • name: duration
    • metric: 0.21276339292526245
    • unit: s
    • diff: 0.03077014088630675
    • improvements: 8
    • regressions: 23