Comparison of different PEFT techniques of NeuralMonarch.
Zain ul Abideen
abideen
AI & ML interests
Large language models
Organizations
Comparison of different PEFT techniques of NeuralMonarch.
Fine-tunes/Merges of Coding Models.
DPO'd version of Gemma-2b and Gemma-7b
Playing around with different LLM Merges (Slerp, Linear, Ties, Dare, Task Arithmetic).
Models and Code of different MoE merges.
TinyLlama-1.1B model aligned on Intel's Orca dataset. Comparison of DPO/IPO/KTO.