Hugo Koubbi, Louis Hernandez, Matthieu Boussard, Small Lora, Large Lora
We lifted 2 functions out of this paper's own repositories and ran 2 of them in a sandbox. "Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results.
| Repository | Role | Ran |
|---|---|---|
| HugoKoubbi/Transformers-2024-LoRA | canonical | 2 of 2 |
| Function | Status | Where it lives |
|---|---|---|
| formatting_func | Ran | HugoKoubbi/Transformers-2024-LoRA/Llama2_Experiments/code/finetuning_llama2.py pointer only (licence: NONE) · get_code("ead8df8503bc811b") |
| process_prompt | Ran | HugoKoubbi/Transformers-2024-LoRA/Llama2_Experiments/code/finetuning_llama2.py pointer only (licence: NONE) · get_code("713c3e4a1a9bfdd2") |
Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.
Low-Rank Adaptation (LoRA) is the dominant parameter-efficient fine-tuning method due to its favorable compute-performance trade-off, yet it suffers from catastrophic forgetting. We study forgetting through a tractable mean-field selfattention toy model, where tokens evolve as an interacting particle system and LoRA acts as a low-rank perturbation. Using tools from partial differential equations and dynamical systems, we characterize regimes suggesting a phase transition between forgetting and non-forgetting behavior. We show that one phase transition appears with respect to the norm of the perturbation, and the other with respect to the depth of the Transformers. We further bound the time-to-deviation in terms of the perturbation size and spectral quantities, and corroborate the predicted trends with experiments and exploratory analyses on real models under LoRA fine-tuning.
The same record, over MCP at https://syntology.ai/mcp:
get_harvested_code_for_paper("2402.15415")
get_code_for_paper("2402.15415")
have("2402.15415")
Connect an agent — have() is free.