Top all-time · #11693
Alex Strick van Linschoten
Personal blog · mlops.systems/postsaidata
Alex Strick van Linschoten's personal blog at mlops.systems covers ai and data. Since 2022, 6 different Hacker News readers have linked it 8 times, most often “How to think about creating a dataset for LLM fine-tuning evaluation”. No feed was found, so read it on the site itself.
Too few sampled comments (3) to say how often HN recommends it.?
Why HN links it
The blog covers MLOps topics such as LLM finetuning and evaluation. Its three comments discuss finetuning services, a tool idea and LLM prompting, and say nothing about the writing.
Summarised by Claude from a sample of 3 HN comments that link here. Each comment is marked recommending, critical or neutral (citing without judging); how this works.
Most-linked posts?
- How to think about creating a dataset for LLM fine-tuning evaluation2 people · 2 mentions · first linked 2024-06-27
What Hacker News says?
OpenpPipe - https://openpipe.ai/ - is probably the service that most closely resembles what you’re asking for, but I found the evals weren’t really what I wanted — i.e. following my custom evaluation criteria — so you probably will end up having to do that yourself anyway. But for the finetuning, they’re all somewhat the same. Predibase and OpenPipe are two good options for that. Predibase has more base models for y…
Comment linking /posts/2024-06-17-one-click-finetuning.html
on “My finetuned models beat OpenAI's GPT-4”Submitted My finetuned models beat OpenAI's GPT-4 as a story · 414 points
HN discussionThis is part of a dream of a tool I would like: A relational crawler on a particular subject with nuanced, opaque, seemingly-temporally-unrelated connections that show a particular MIC conduction of acts:: "Follow all the congress members who have been a part of a particular committee, track their signatory/support for particular ACTs that have been passed, and look at their investment history from open data, quiver…
Comment linking /
on “Show HN: Turn any website into a knowledge base for LLMs”Submitted How to think about creating a dataset for LLM fine-tuning evaluation as a story · 139 points
HN discussion> You can see this with even just very large prompts, the larger the context the worse the quality, despite model developers claiming ever larger context windows. This is precicely the problem I am attempting to address. The point is that I want to be able to frame the way a particular complex iterative query discovery can be accomplished where I utilized already baked frames of how I want to address the problem, si…
Comment linking /posts/2024-06-25-evaluation-finetuning-manual-dataset.html
on “Ask HN: Discernment Lattice [Prompting]”Submitted Storing Bytes: what data serialisation is and why you need it as a story · 1 points
HN discussionSubmitted How my pet cat taught me a lesson about validation data for image classification as a story · 1 points
HN discussion