Yi: Open Foundation Models by 01.AI · HackerPapers discussions

Yi: Open Foundation Models by 01.AI — mentioned by numpad0

Mon, 13 May 2024 11:32:35 GMT

Earlier Yi paper indicates it's trained with less than 25% Chinese dataset, contrast to GPT-3 which was 93% English[1][2]. Is that a bug or could there be something inherent to current LLM architecture - like dataset must be 90%+ English to not fall apart? 1: https://arxiv.org/html/2403.04652v1 2: https://github.com/openai/gpt-3/blob/master/dataset_statisti...

Paper and discussion history. A mention is not an endorsement.

Yi: Open Foundation Models by 01.AI — HN submission

Sun, 10 Mar 2024 15:12:30 GMT

Yi: Open Foundation Models by 01.AI

205 points; 81 discussion replies on HN. Paper and source history. A submission is not an endorsement.

Yi: Open Foundation Models by 01.AI — HN submission

Fri, 08 Mar 2024 16:11:01 GMT

Yi: Open Foundation Models

1 point; 0 discussion replies on HN. Paper and source history. A submission is not an endorsement.