| Model Type | | text-to-text, decoder-only, language |
|
| Use Cases |
| Areas: | | Content Creation and Communication, Research and Education |
|
| Applications: | | Text Generation, Chatbots and Conversational AI, Text Summarization, NLP Research, Language Learning Tools, Knowledge Exploration |
|
| Limitations: | | Training Data quality and scope, Context and Task Complexity, Language Ambiguity and Nuance, Factual Accuracy, Common Sense |
|
| Considerations: | |
|
| Supported Languages | |
| Training Details |
| Data Sources: | | Web Documents, Code, Mathematics |
|
| Data Volume: | | 8 trillion tokens for 9B model |
|
| Methodology: | |
| Hardware Used: | |
|
| Safety Evaluation |
| Methodologies: | | structured evaluations, internal red-teaming |
|
| Findings: | | acceptable thresholds for internal policies |
|
| Risk Categories: | | Text-to-Text Content Safety, Text-to-Text Representational Harms, Memorization, Large-scale harm |
|
|
| Responsible Ai Considerations |
| Fairness: | | LLMs trained on large-scale, real-world text data can reflect socio-cultural biases. |
|
| Transparency: | | Model card present with details on architecture, capabilities, limitations, and evaluation processes. |
|
| Mitigation Strategies: | | Mechanisms and guidelines for content safety provided for developers. |
|
|
| Input Output |
| Input Format: | |
| Accepted Modalities: | |
| Output Format: | | Generated English-language text |
|
|
