> You'll get much further if you do one or two simple passes on your writing, and then pass what you've written around and ask for feedback.
I've found that the more time you spend "intellectualizing" your own thoughts, it'll become harder to back down, because you get so invested in your ideas.
> but of course our lives are finite and we have to make choices that might limit our development of any skill.
This is a key point in Four Thousand Weeks (https://www.goodreads.com/en/book/show/54785515). For our hobbies, it'd be totally fine (and even preferable) to not worry about percentiles at all.
We're working on LLM Engine (https://llm-engine.scale.com) at Scale, which is our open source, self-hostable framework for open source LLM inference and fine-tuning. We have similar findings to Replicate: Llama 2 70B can be comparable to GPT 3.5 price, etc. Would be great to discuss this further!
> You'll get much further if you do one or two simple passes on your writing, and then pass what you've written around and ask for feedback.
I've found that the more time you spend "intellectualizing" your own thoughts, it'll become harder to back down, because you get so invested in your ideas.