Starcoderbase Triviaqa is an open-source language model by vwxyzjn. Features: 15.5b LLM, VRAM: 62.1GB, License: bigscience-openrail-m, Code Generating, LLM Explorer Score: 0.09.
May generate incorrect or misleading answers, May copy answers from the training data verbatim, May generate language that is hateful or discriminatory, May generate offensive language
Considerations:
Answers should be validated through external sources.
Supported Languages
en (English)
Training Details
Data Sources:
TriviaQA
Methodology:
Fine-tuned using reinforcement learning via TRL's `TextEnvironment`.
Responsible Ai Considerations
Mitigation Strategies:
Disparities between the data contributors and users should inform developers in assessing appropriate use cases. Further research is needed to attribute model generations to sources in the training data.
Note: green Score (e.g. "73.2") means that the model is better than vwxyzjn/starcoderbase-triviaqa.
Rank the Starcoderbase Triviaqa Capabilities
🆘 Have you tried this model? Rate its performance. This feedback would greatly assist ML community in identifying the most suitable model for their needs. Your contribution really does make a difference! 🌟
Instruction Following and Task Automation
Factuality and Completeness of Knowledge
Censorship and Alignment
Data Analysis and Insight Generation
Text Generation
Text Summarization and Feature Extraction
Code Generation
Multi-Language Support and Translation
What open-source LLMs or SLMs are you in search of? 53972 in total.