Phi 3 Mini 4K Instruct Cinder With 16bit GGUF is an open-source language model by Josephgflowers. Features: 3.8b LLM, VRAM: 7.7GB, Context: 4K, License: mit, Quantized, Instruction-Based, LLM Explorer Score: 0.12.
Models are not specifically designed for all downstream purposes and have limitations in scope.
Considerations:
Developers should evaluate use cases for safety and fairness before deployment.
Supported Languages
en (primary)
Training Details
Data Volume:
3.3 trillion tokens
Methodology:
Supervised fine-tuning (SFT) and Direct Preference Optimization (DPO)
Context Length:
4096
Training Time:
7 days
Hardware Used:
512 H100-80G GPUs
Model Architecture:
dense decoder-only Transformer
Responsible Ai Considerations
Fairness:
These models can potentially behave in ways that are unfair, unreliable, or offensive. They may over- or under-represent groups of people or reinforce stereotypes.
Transparency:
Developers should apply responsible AI best practices and are responsible for ensuring that a specific use case complies with relevant laws and regulations.
Accountability:
Developers are responsible for the use of the model and ensuring compliance with relevant laws.
Mitigation Strategies:
Despite safety post-training, developers should assess outputs for context and use available safety classifiers.
Input Output
Input Format:
Chat format with roles and system instructions.
Accepted Modalities:
Text
Output Format:
Text response to input
Performance Tips:
For optimal use, ensure proper formatting of chat prompts.