22 Matching Annotations
  1. Dec 2023
    1. Books that bust bubbles

      Amazing book recommendations that are almost like "quake" books!

    2. compliment

      *complement?

  2. May 2023
    1. All Roads Lead to Rome: The Machine Learning Job Market in 2022

      Oh! This article that I think I say around the time it released is so inspiring - especially for AI humanoid robots.

    1. Google said in a blog post about PaLM 2 that the model uses a “new technique” called “compute-optimal scaling.” That makes the LLM “more efficient with overall better performance, including faster inference, fewer parameters to serve, and a lower serving cost.”

      This technique is simply not falling for the "Chinchilla Trap" that Andrej of OpenAI (formerly Tesla AI etc) mentions in this tweet here. He also explains this in the comments. https://twitter.com/karpathy/status/1654892810590650376?s=20

  3. Mar 2023
  4. Feb 2023
    1. $(DOCKER) $(CROSS)-objcopy

      Needs to be "arm-none-eabi-objcopy"

    2. SYSTICK->LOAD = ticks - 1; SYSTICK->VAL = 0; SYSTICK->CTRL = BIT(0) | BIT(1) | BIT(2);

      This is explained in a doc titled "STM32 Cortex®-M4 MCUs and MPUs programming manual - Programming manual" > 4.5.1 + 4.5.2 + 4.5.3

      https://www.st.com/resource/en/programming_manual/pm0214-stm32-cortexm4-mcus-and-mpus-programming-manual-stmicroelectronics.pdf

      SYSTICK->CTRL = BIT(0) | BIT(1) | BIT(2); --> BIT(0): Counter enabled BIT(1): Counting down to zero to asserts the SysTick exception request. BIT(2): Clock Source = Processor clock (AHB)

    3. $(TARGET).bin

      This needed to be "&<" for me

    4. pin

      Needs to be "led"

    5. pin

      Needs to be "led"

    6. 2**2

      My value is 2**1. Not sure if an issue..

    7. 00000058

      My value is 000002. Maybe not an issue

    8. ...

      Once again:

      After 1 .vectors, I also have 2 .comment and 3 .ARM.attributes. That's probably from updated packages I downloaded vs CPQ's versions

    9. ...

      After 3 .vectors, I also have 4 .comment and 5 .ARM.attributes. That's probably from updated packages I downloaded vs CPQ's versions

    10. to

      *no

  5. Jan 2023
    1. Wipe On, Wipe Off: Make Your Own Rain Repellent

      If safe, this would be an awesome project for kids. Learn that the products in our world are made by humans no smarter than ourselves

    1. Perplexity is a measurement of randomness in a sentence. If a sentence is constructed or uses words in a way that surprises the app, then it will score higher in perplexity. Tian said that he used the free and open source GPT-2 to help train his app for this metric.Burstiness is the quality of overall randomness for all the sentences in a text. For example, human writing tends to have sentences that vary in complexity. Some are simple. Some can give James Joyce a run for his money. Bots, on the other hand, tend to generate sentences that are relatively low in complexity, throughout the entire text.

      TIL - also interesting that humans will probably write more perplex and more bursty in response