A continuation of OpenHermes 2 model, trained on additional code datasets. Potentially the most interesting finding from training on a good ratio (est. of around 7-14% of the total dataset) of code instruction was that it has boosted several non-code benchmarks, including TruthfulQA, AGIEval, and GPT4All suite. It did however reduce BigBench benchmark score, but the net gain overall is significant.
模态
上下文
4K
发布时间
2023年11月20日
知识截止日期
2023年9月
目前无法通过 EU 区域内路由使用。 区域内路由文档