H2O-Danube-1.8B Technical Report · HackerPapers discussions

H2O-Danube-1.8B Technical Report — mentioned by nickpsecurity

Thu, 30 Apr 2026 20:38:02 GMT

On changing the training mix, H20 did that with Danube in 2024: https://arxiv.org/pdf/2401.16818 With those results, I would've already done that in any models I got to train. There's also the principle that the LLM's are often better at what they saw last in their training set. That also justifies putting more logic, code, and math in at the end for an analytical or coding model. So, a few precedents for that technique already.

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

H2O-Danube-1.8B Technical Report — HN submission

Sun, 07 Apr 2024 18:49:20 GMT

H2O-Danube-1.8B Technical Report

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

H2O-Danube-1.8B Technical Report — HN submission

Fri, 05 Apr 2024 19:24:28 GMT

H2O-Danube-1.8B: Technical Report

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