Skip to content

Latest commit

 

History

History
60 lines (49 loc) · 2.67 KB

File metadata and controls

60 lines (49 loc) · 2.67 KB

Enum: AdapterType

URI: nexus:AdapterType

Permissible Values

Value Meaning Description
LORA None Low-rank adapters, or LoRAs, are a fast way to give generalist large language...
ALORA None Activated LoRA (aLoRA) is a low rank adapter architecture that allows for reu...
X-LORA None Mixture of LoRA Experts (X-LoRA) is a mixture of experts method for LoRA whic...

Slots

Name Description
hasAdapterType The Adapter type, for example: LORA, ALORA, X-LORA

Identifier and Mapping Information

Schema Source

LinkML Source

Details ```yaml name: AdapterType from_schema: https://ibm.github.io/ai-atlas-nexus/ontology/ai-risk-ontology rank: 1000 permissible_values: LORA: text: LORA description: Low-rank adapters, or LoRAs, are a fast way to give generalist large language models targeted knowledge and skills so they can do things like summarize IT manuals or rate the accuracy of their own answers. LoRA reduces the number of trainable parameters by learning pairs of rank-decompostion matrices while freezing the original weights. This vastly reduces the storage requirement for large language models adapted to specific tasks and enables efficient task-switching during deployment all without introducing inference latency. LoRA also outperforms several other adaptation methods including adapter, prefix-tuning, and fine-tuning. See arXiv:2106.09685 ALORA: text: ALORA description: Activated LoRA (aLoRA) is a low rank adapter architecture that allows for reusing existing base model KV cache for more efficient inference, unlike standard LoRA models. As a result, aLoRA models can be quickly invoked as-needed for specialized tasks during (long) flows where the base model is primarily used, avoiding potentially expensive prefill costs in terms of latency, throughput, and GPU memory. See arXiv:2504.12397 for further details. X-LORA: text: X-LORA description: Mixture of LoRA Experts (X-LoRA) is a mixture of experts method for LoRA which works by using dense or sparse gating to dynamically activate LoRA experts.
</details>