Style Vectors for Steering Generative Large Language Model
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2024
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866929231950249984 |
|---|---|
| author | Konen, Kai Jentzsch, Sophie Diallo, Diaoulé Schütt, Peer Bensch, Oliver Baff, Roxanne El Opitz, Dominik Hecking, Tobias |
| author_facet | Konen, Kai Jentzsch, Sophie Diallo, Diaoulé Schütt, Peer Bensch, Oliver Baff, Roxanne El Opitz, Dominik Hecking, Tobias |
| contents | This research explores strategies for steering the output of large language models (LLMs) towards specific styles, such as sentiment, emotion, or writing style, by adding style vectors to the activations of hidden layers during text generation. We show that style vectors can be simply computed from recorded layer activations for input texts in a specific style in contrast to more complex training-based approaches. Through a series of experiments, we demonstrate the effectiveness of activation engineering using such style vectors to influence the style of generated text in a nuanced and parameterisable way, distinguishing it from prompt engineering. The presented research constitutes a significant step towards developing more adaptive and effective AI-empowered interactive systems. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2402_01618 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Style Vectors for Steering Generative Large Language Model Konen, Kai Jentzsch, Sophie Diallo, Diaoulé Schütt, Peer Bensch, Oliver Baff, Roxanne El Opitz, Dominik Hecking, Tobias Computation and Language This research explores strategies for steering the output of large language models (LLMs) towards specific styles, such as sentiment, emotion, or writing style, by adding style vectors to the activations of hidden layers during text generation. We show that style vectors can be simply computed from recorded layer activations for input texts in a specific style in contrast to more complex training-based approaches. Through a series of experiments, we demonstrate the effectiveness of activation engineering using such style vectors to influence the style of generated text in a nuanced and parameterisable way, distinguishing it from prompt engineering. The presented research constitutes a significant step towards developing more adaptive and effective AI-empowered interactive systems. |
| title | Style Vectors for Steering Generative Large Language Model |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2402.01618 |