MindBranches
@MindBranches
Followers
14,587
Following
1,381
Media
1,965
Statuses
3,502
AI enhanced diagrams to help you understand complex concepts quickly.
United States
Joined March 2023
Don't wanna be here?
Send us removal request.
Explore trending content on Musk Viewer
Marcellus Williams
• 685504 Tweets
River
• 211616 Tweets
LINGLING PR MYCHERIE AMOUR
• 114963 Tweets
Ghost of Yotei
• 110291 Tweets
#CathyDollxLingOrm
• 105973 Tweets
Quito
• 91239 Tweets
Vidal
• 75449 Tweets
#RRBNTPC_VACANCY_INCREASE
• 69719 Tweets
Padres
• 66581 Tweets
Colo Colo
• 50579 Tweets
Gallardo
• 44254 Tweets
#ซองแดงแต่งผี
• 29303 Tweets
Dodgers
• 29183 Tweets
スペック
• 28424 Tweets
Colidio
• 27766 Tweets
पंडित दीनदयाल उपाध्याय
• 25289 Tweets
#CKFragranceEndorserZee
• 23194 Tweets
シーザー
• 20754 Tweets
Bulacan
• 19717 Tweets
モンハン
• 19491 Tweets
Rojas
• 14955 Tweets
TRIPLE PLAY
• 14361 Tweets
#DAY6INBKK2024
• 13694 Tweets
एकात्म मानववाद
• 12232 Tweets
コリンク
• 10722 Tweets
予告番組
• 10672 Tweets
Pinned Tweet
Introducing my new book “The Art of Crafting Prompts: An Interactive Guide To Mastering Chat GPT With MindBranches”
Quickly learn:
• Strategies for thinking with AI
• Keys to effective prompting
• How to make MindBranches
Packed with exercises and visuals
(Link in bio)
4
7
50
Open AI released a great Prompt Engineering guide (link in comments).
Here is a summary of their 6 strategies for getting better results when prompting GPT-4
54
1K
7K
When you ask ChatGPT to act as an expert it significantly improves your outputs.
Here are some expert roles to try.
Just start your prompt with: “As an expert [pick a role from the list] (enter the rest of your prompt…)”
9
113
1K
GPT-4o knows a lot about various well known frameworks.
You can include the frameworks below in your prompts.
Example: “Use Porter's Five Forces
Focusing on [specific industry]”
10
144
1K
Here’s a great way to use ChatGPT to learn more about technology.
Pick a topic from the MindBranch below and insert it into the following prompt:
"Act as an expert educator in the field of [topic]. I'm seeking a comprehensive understanding of this subject and would like you to
22
113
605
20
48
486
Link to OpenAI’s full Prompt Engineering Guide:
2
94
427
My new favorite prompting strategy:
“Chain of Thought Leaders Prompting”
It works by leveraging the thousands of books that Large Language Models have access to from their training data to provide the most useful insights from the most important thought leaders on any given
16
83
393
@george__mack
This is a great compilation of useful razors and rules by
@george__mack
Here they are categorized for quick reference:
3
59
380
@sama
Looks great!
I asked the *new* version of GPT-4 turbo what its initial instructions are and it provided the following info:
28
31
359
This multi-step prompt is a quick way to unlock the knowledge from all the books that large language models were trained on.
I call it “Chain of Thought Leaders Prompting”
Instructions: enter each of these as a separate prompt after you have received a response to the previous
3
29
288
@edleonklinger
@SteveMoraco
Between the eye tracking of Vision and heart monitor on the Apple Watch I wouldn’t be surprised to see a version released at some point with full Neurofeedback
4
36
280
@lexfridman
@elonmusk
@neuralink
Brief Summary of Lex Fridman Podcast
#438
"Elon Musk: Neuralink and the Future of Humanity"
18
35
271
@AndrewYNg
@rm_rafailov
@archit_sharma97
@ericmitchellai
@StefanoErmon
@chrmanning
@chelseabfinn
Here’s a summary of the research paper “Direct Preference Optimization:
Your Language Model is Secretly a Reward Model”
4
28
261
Microsoft Copilot is inserting Ads as sources for its outputs. 👀
I wonder if Google Gemini will eventually do the same.
Maybe it’s just me, but I don’t like the idea of my AI outputs being determined by the highest bidder.🤷♂️
71
30
255
@javilopen
How much art would there be if every artist listened to their critics?
Here are some examples of criticism towards new art forms throughout the ages:
21
49
232
@janleike
@AnthropicAI
This will add a ton of value to the Anthropic team, similar to how the Golden Gate Bridge adds value to San Francisco. He will help to cut through the mists of AI safety research to provide a path forward over the sea of AI risk.
10
8
205
13
20
202
7
4
191
GPT-o1 preview is FAST!
I asked it to generate a white paper on generative AI use cases for businesses and it did it almost instantly, complete with sources.
5
19
183
This is blowing my mind. I’m pretty sure OpenAI just did a major behind the scenes update.
Start a prompt with: “Reason fully before answering. Assume the physics of our world.”
Then try to stump it (4o nailed this example prompt):
“Reason fully before answering. Assume the
16
16
149
@tferriss
Here’s a summary of the book “Secrets of Power Negotiating” by Roger Dawson including specific strategies for improving negotiations:
2
20
133
It’s an honor to be acknowledged by one of my favorite people on 𝕏.
@MindBranches
are carefully crafted diagrams that use AI technologies to help you quickly digest large amounts of useful information.
Like Brian, I strive to keep my followers ahead of the curve.
I am a massive fan of
@MindBranches
.
MindBranches, by Teamonics. uses GPT-4 AI present information and complex concepts quickly.
I think all AI platforms should use this as a visual tracking interface.
Just brilliant!
Link:
A MindBranch specimen:
11
22
157
13
26
130
I’m convinced people who aren’t impressed by generative AI just haven’t tried using it enough.
Here are some things to try:
1
20
130
@LinusEkenstam
I made this GPT to show people new to GPT what it can do. Give it a topic and it generates advanced prompt examples and outputs that users rank on an impressiveness scale of 1 to 10 and keeps trying until you are thoroughly impressed.
10
4
127
@soncharm
I prefer using AI to gain a deeper appreciation of the original.
I asked GPT-4V to tell me about the painting.
Output in MindBranches:
10
7
122
@itsurboyevan
@OpenAI
Code interpreter is the 🐐GPT
Ask it to run a 10k iteration Monte Carlo next time you need to estimate anything with multiple variables.
Prompts I used laid out in MindBranches & the outputs:
5
6
116
@Scobleizer
There are plenty of business use cases once leaders wrap their heads around what LLM’s can do.
5
10
116
Forbes just released their AI 50 list.
Here’s a categorized list of the companies, including their current funding amounts:
5
27
110
Chat GPT is the ultimate thinking partner and the more it knows about you the better it works.
Give it a bunch of context about yourself using the relevant personal context prompts in the first image, then run one of the dialogue prompts in the second image.
Do this right and
5
15
109
@skirano
Having GPT generate prompts and then asking it to break them down by components and categories is also very useful. Here’s an example prompt I had it generate broken down by components and displayed in MindBranches. (Easy to see where you can adjust the prompt for your use case)
4
13
108
@OpenAI
If you evenly spaced out your prompts you can now ask GPT-o1-preview a question every 3.36 hours.
18
3
176
@sama
What a year indeed!
Great work by you and the OpenAI team ❤️
Here’s a summary of OpenAI’s accomplishments over the past year according to GPT-4:
4
10
101
@thatroblennon
Now for the real magic:
Prompt:
Write a similar prompt to this, but make it related to business and innovative ideas rather than biology or biometric design:
[paste entire NASA prompt]
(Spoiler, the output is amazing!)
2
6
100
@Yampeleg
Summary of the research paper: “Soaring from 4K to 400K: Extending LLM’s Context with Activation Beacon”
6
9
96
@svpino
Hopefully the algorithm is continually improved to promote quality over click bait.
Anyways, here’s a list of reasons to prioritize quality content over clickbait:
2
11
99
@OpenAI
Here is an overview of Sora, OpenAI’s brand new text-to-video model:
(Info summarized from OpenAI’s website)
4
20
93
I think that we are barely scratching the surface of what GPT-4 is capable of. Here is a 60 part prompt for an online business idea with output laid out in MindBranches. (PDF on my website)
#HustleGPT
4
15
92
@OpenAI
Summary of OpenAI Leadership Announcement.
Sam Altman departs as CEO and leaves the board of directors.
1
6
92
Is writing in all lowercase an internal openAI thing so they know it’s not GPT generated?
9
6
88
@svpino
This is super helpful for someone getting started with VS Code.
Here’s an overview of your preferred VS code extensions with added summaries of what they are used for:
1
13
83
@emollick
Wow, my initial reaction to this is how disorienting it’s going to be as LLM hallucinations approach 0%. I expect I’ll have a few more bouts of “three sleepless nights” to look forward to.
3
0
82
@rowancheung
What a week for AI😵💫
Thanks
@rowancheung
for a fantastic overview. I put together a summary of your weekly AI updates:
5
15
80
@Yampeleg
Even the current models are way more capable than what’s available to most. I had very early GPT-4 beta access (before they put up a bunch of guardrails in the system prompt) and it was amazing. They could just take off the guardrails and call it GPT-5.
3
3
81
Putting the finishing touches on my guide to crafting prompts.
(It also includes a bonus section that teaches you how to make MindBranches!)
6
6
82
@AdrianDittmann
Great overview of the 𝕏 algo changes by
@AdrianDittmann
.
I put together a summary:
8
27
78
@george__mack
Great concept.
Here are activities that get you off of the hedonic treadmill and on to the hedonic escalator offering sustained or increasing satisfaction over time:
1
17
80
3
2
73
@emollick
Fantastic article and exciting research on the implications of AI on work.
I put together a summary of your article:
2
9
73
I cringe every time I see one of these words in my GPT outputs.
Words Chat GPT Over Uses:
(Did I miss any?)
16
7
73
@NorthstarBrain
They spent 100 million dollars tuning a large language model to generate the most useful outputs possible then ruin those outputs by stuffing them with ads.
If they want to run ads, they should have them display separately on the page to maintain the integrity of the outputs.
1
1
73
@lexfridman
Humans are incredible problem solvers.
Here is an overview of our advanced problem solving capabilities:
6
13
70
6
3
70
Meta’s new Segment Anything Model 2 (SAM 2) is incredible!
I fed it this video and it instantly was able to track Craig’s complex movements and add text overlays.
1
11
66
Just finished watching the 1 hour “Intro to Large Language Models”video by
@karpathy
This video is a must watch for anyone interested in LLM’s.
I put together a summary of the video below.
Full Video:
4
10
68
