Experiments & Evaluations
AI experiments, benchmarks, and evaluation projects (18 projects).
This listing is generated. For the hand-curated version, see the Experiments & Evaluations index repository →
Showing projects
GitHubRepositoryAgent-Handover-Demodanielrosehill/Agent-Handover-DemoModel for simple handover management (via docs!) between AI agentsView projectGitHubRepositoryAI-Agent-UNdanielrosehill/AI-Agent-UNExperimentary implementation of a model real life political body populated by representative AI agentsView projectGitHubRepositoryAI-Assistant-Templatedanielrosehill/AI-Assistant-TemplateTemplate repository for version controlling AI assistant configurations (and associated elements like context data)View projectGitHubRepositoryAI-Dev-Prompts-Exampledanielrosehill/AI-Dev-Prompts-ExampleExamples of AI prompts for code generation (Python GUI) and some notes on emerging "best practices"View projectGitHubRepositoryBias-Censorship-Eval-Testsdanielrosehill/Bias-Censorship-Eval-TestsSome loosely organised experiments intended to probe the levels of political censorship inherent in LLMsView projectGitHubRepositoryCrying-Baby-Audio-Scrubdanielrosehill/Crying-Baby-Audio-ScrubExperiment to see if crying baby audio can be scrubbed with audio inpainting (use-case: WIP parenting and content creation!)View projectGitHubRepositoryGHG-EBITDA-Correlationsdanielrosehill/GHG-EBITDA-CorrelationsStreamlit data visualisation examining correlation between emissions & profitabilityView projectGitHubRepositoryLLM-Evaluation-Promptsdanielrosehill/LLM-Evaluation-PromptsA few prompts that I am storing in a repo for the purpose of running controlled experiments comparing and benchmarking different LLMs for defined use-casesView projectGitHubRepositoryLLM-Experiment-Notebookdanielrosehill/LLM-Experiment-NotebookExperiments in evaluating various prompting strategies and LLM performance generallyView projectGitHubRepositoryLLM-Long-Codegen-Testdanielrosehill/LLM-Long-Codegen-TestExperiment to evaluate the ability of code-gen LLMs to generate a long continuous (single) outputView projectGitHubRepositoryLocal-ASR-STT-Benchmarkdanielrosehill/Local-ASR-STT-BenchmarkQuick evaluation to find the best STT model in Speech Note (Ubuntu) for my HWView projectGitHubRepositoryLong-AI-Prompting-Experimentdanielrosehill/Long-AI-Prompting-ExperimentExperiment testing the "added value" of long context-dense prompts (versus casual converastional prompts) and the value of a voice "prompt optimisation" agentView projectGitHubRepositoryLong-Form-Audio-Evaldanielrosehill/Long-Form-Audio-EvalSingle shot STT benchmark for long from audioView projectGitHubRepositoryOne-Prompt-AI-Bookdanielrosehill/One-Prompt-AI-BookExperiment - generating a book using a single promptView projectGitHubRepositoryOSINT-Missile-Intelligence-Agentdanielrosehill/OSINT-Missile-Intelligence-AgentSnap reports about causes and quantitiative info about hostile missile launchesView projectGitHubRepositorySTT-Voice-Note-Evaluationdanielrosehill/STT-Voice-Note-EvaluationView projectGitHubRepositoryTest-Markdown-Docsdanielrosehill/Test-Markdown-DocsA few formatted markdown files for testing the real-time markdown rendering support of various knowledge management, note-taking and wiki toolsView projectGitHubRepositoryTest-System-Promptsdanielrosehill/Test-System-PromptsA collection of system prompts organised by capabilities for testing out various agent/assistant platformsView project