Dirk Groeneveld
First author of Ai2's OLMo paper; contributor to OLMo 2, OLMoE, and the Dolma corpus
Dirk Groeneveld is first author of the 2024 OLMo paper, which released Ai2's open language model together with its training data and its training and evaluation code. In the OLMo 2 paper, where Groeneveld is listed with an Ai2 affiliation, Groeneveld is credited with base-model development, including training and data curation. The OLMoE paper credits Groeneveld with advising on pretraining, especially stability and throughput improvements, and Groeneveld is also a co-author of the Dolma pretraining corpus paper.1234
Work
OLMoOpen weights
First author of the OLMo paper (2024); credited in OLMo 2 with base-model development, including training and data curation; credited in OLMoE with advising on pretraining stability and throughput.123
DolmaOpen data
Co-author of the Dolma paper, which describes a three-trillion-token English corpus released for research on language-model pretraining.4
Other documented roles
Roles the sources document but that are not confirmed as current.
- Base-model developer on OLMo 2, listed with an Ai2 affiliation (2024), Ai2 (Allen Institute for AI)2
This profile covers public professional work only, from the sources below. It is not an endorsement or a ranking. To correct or remove it, use the correction form.