Salvato in:
| Autori principali: | , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | https://arxiv.org/abs/2503.23298 |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866912300830556160 |
|---|---|
| author | Wang, Jiachuan Di, Shimin Tang, Tianhao LI, Haoyang Ng, Charles Wang-wai Zhou, Xiaofang Chen, Lei |
| author_facet | Wang, Jiachuan Di, Shimin Tang, Tianhao LI, Haoyang Ng, Charles Wang-wai Zhou, Xiaofang Chen, Lei |
| contents | Emergence, the phenomenon of a rapid performance increase once the model scale reaches a threshold, has achieved widespread attention recently. The literature has observed that monosemantic neurons in neural networks gradually diminish as the model scale increases. Subsequently, Learning From Emergence is proposed to actively inhibit monosemantic neurons in relatively small neural networks (e.g., BERT and Swin-Transformer) for promoting model performance with fine-tuning. However, to ultimately achieve emergence, it is demanding to support the monosemantic neuron inhibition in the pretraining phase of large-scale models. Thus, this work further pushes the boundary of this research direction to be Learning Towards Emergence (L2E) and enables the training and validating of the impact of inhibiting monosemantic neurons on larger pre-trained neural networks (e.g., Pythia-70M, 410M, and 2.8B). More specifically, to bridge the gap in current research, we first conduct experiments on models of various scales (up to 6.9B) to validate the monosemantic ideas. Then, we present a novel method L2E to address the inefficient monosemantic neuron retrieval and ineffective monosemantic neuron inhibition when existing methods are applied in the pretraining phase of large-scale models. It employs an adjustable thresholding technique for efficient neuron retrieval, incorporates a False Killing Rate metric to assess inhibition effects, and proposes a regularization-style inhibition approach, which addresses the limitations of previous approaches in both efficiency and effectiveness. Experimental results demonstrate the effectiveness of L2E's monosemantic neuron inhibition and its efficiency in implementation with large-scale models. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2503_23298 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Learning Towards Emergence: Paving the Way to Induce Emergence by Inhibiting Monosemantic Neurons on Pre-trained Models Wang, Jiachuan Di, Shimin Tang, Tianhao LI, Haoyang Ng, Charles Wang-wai Zhou, Xiaofang Chen, Lei Emerging Technologies Emergence, the phenomenon of a rapid performance increase once the model scale reaches a threshold, has achieved widespread attention recently. The literature has observed that monosemantic neurons in neural networks gradually diminish as the model scale increases. Subsequently, Learning From Emergence is proposed to actively inhibit monosemantic neurons in relatively small neural networks (e.g., BERT and Swin-Transformer) for promoting model performance with fine-tuning. However, to ultimately achieve emergence, it is demanding to support the monosemantic neuron inhibition in the pretraining phase of large-scale models. Thus, this work further pushes the boundary of this research direction to be Learning Towards Emergence (L2E) and enables the training and validating of the impact of inhibiting monosemantic neurons on larger pre-trained neural networks (e.g., Pythia-70M, 410M, and 2.8B). More specifically, to bridge the gap in current research, we first conduct experiments on models of various scales (up to 6.9B) to validate the monosemantic ideas. Then, we present a novel method L2E to address the inefficient monosemantic neuron retrieval and ineffective monosemantic neuron inhibition when existing methods are applied in the pretraining phase of large-scale models. It employs an adjustable thresholding technique for efficient neuron retrieval, incorporates a False Killing Rate metric to assess inhibition effects, and proposes a regularization-style inhibition approach, which addresses the limitations of previous approaches in both efficiency and effectiveness. Experimental results demonstrate the effectiveness of L2E's monosemantic neuron inhibition and its efficiency in implementation with large-scale models. |
| title | Learning Towards Emergence: Paving the Way to Induce Emergence by Inhibiting Monosemantic Neurons on Pre-trained Models |
| topic | Emerging Technologies |
| url | https://arxiv.org/abs/2503.23298 |