Model Type | |
Use Cases |
Areas: | |
Applications: | assistant-like chat, natural language generation tasks |
|
Primary Use Cases: | Instruction tuned models for dialogue, Pretrained models for various text tasks |
|
Limitations: | Use violating laws or policies, Use in non-English languages without compliance |
|
|
Supported Languages | |
Training Details |
Data Sources: | publicly available online data |
|
Data Volume: | |
Methodology: | supervised fine-tuning (SFT) and reinforcement learning with human feedback (RLHF) |
|
Context Length: | |
Hardware Used: | Meta's Research SuperCluster, H100-80GB GPU |
|
Model Architecture: | auto-regressive language model with an optimized transformer architecture |
|
|
Safety Evaluation |
Methodologies: | red teaming, adversarial evaluations |
|
Findings: | reduced residual risks with safety tools, improved refusal rates to benign prompts |
|
Risk Categories: | misinformation, child safety, cybersecurity |
|
Ethical Considerations: | Emphasis on responsible deployment and safety best practices |
|
|
Responsible Ai Considerations |
Accountability: | |
Mitigation Strategies: | Provides tools like Meta Llama Guard 2 and Code Shield for developers |
|
|
Input Output |
Input Format: | |
Accepted Modalities: | |
Output Format: | |
|
Release Notes |
Version: | |
Date: | |
Notes: | Initial release of Llama 3 family models |
|
|
|