OPT
1. Model Introduction
OPT: Open Pre-trained Transformer Language Models is a general-purpose language model trained with autoregressive blank-filling objectives, suitable for various understanding and generation tasks.
Supported Model Weights:
| Model |
|---|
| facebook/opt-125m |
| facebook/opt-350m |
| facebook/opt-1.3b |
| facebook/opt-2.7b |
| facebook/opt-6.7b |
| facebook/opt-13b |
| facebook/opt-30b |
| facebook/opt-66b |
| facebook/opt-iml-1.3b |
| opt-iml-max-1.3b |
Note: The model implementation is available at https://github.com/facebookresearch/metaseq. The OPT-IML (Instruction Meta-Learning) series models demonstrate enhanced instruction-following capabilities through meta-learning techniques.