Veloshnee Naicker

[Veloshnee Naicker @VeloshneeN]

@reachvaldo TopGPT

[Veloshnee Naicker @VeloshneeN]

@Zendesk Make sure to keep the customer in the loop about what's being done to fix their problem. Uncertainty can really ramp up frustration, so it's important to keep them updated regularly, even if you don’t have the best news to share.

[Eric Vaughan @TheGenAICEO]

More coverage on our incredible AI-powered announcement launched at @imagineailive 2024. This idea is exciting customers with ideas we hadn't thought of-- but then that's always the case! https://t.co/WDLX4DYbch

[Ethan Mollick @emollick]

Big day for unexpectedly powerful LLM releases.

Microsoft's open source WizardLM 2 (also note that it used synthetic inputs in training, maybe "running out of data" will not be a big deal): https://t.co/EDq935uguF

Closed source Reka, which is multimodal: https://t.co/k2B81h9vv0

[GFI Software @GFISoftware]

Can a buzzing phone lead to a major data breach? Find out tomorrow in our exclusive webinar: Hidden Risks: How Workplace Distractions Compromise Cybersecurity. Discover the unexpected risks that could impact YOUR business.

Register now 👇

[Veloshnee Naicker @VeloshneeN]

Exciting new features on @poe_platform!Context-aware recommendations let you compare answers across bots,@-mention any bot to bring it into your conversation.Leverage GPT-4,Claude,DALL-E 3 & more in one thread.Explore the strengths of each model & find the best fit for your task!

[MindBranches @MindBranches]

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 topic.

It then uses the power of chain of thought prompting to explore topics more deeply by enhancing context with the wisdom of thought leaders.

Here is the full “Chain of Thought Leaders” prompt series (enter each step as a separate prompt after you have received a response to the previous prompt):

Prompt 1:

Identify the top books by thought leaders in [topic]

Prompt 2:

Provide a comprehensive summary list of the key ideas from each of these books relevant to [topic]. Aim for 5 bullet points for each book.

Prompt 3:

Synthesize the ideas above into actionable insights related to [topic], focusing on their implications and applications.

Prompt 4:

Form a cohesive straightforward narrative that integrates these insights into a comprehensive overview of [topic], addressing its key aspects and implications.

Prompt 5:

Refine the narrative to ensure it is clear and comprehensive. Provide a summary that encapsulates the essential insights and perspectives on [topic].

[Ethan Mollick @emollick]

Pro-tip: remember that if you want Llama 2 to do well on a 50 question math test, pretend you are in Star Trek. To solve 100 questions, pretend you are in a political thriller.

Results like this are why I stopped worrying about optimal prompting too much. https://t.co/ehnBQTfPgq

[Veloshnee Naicker @VeloshneeN]

@bentossell Thank you for this! Bookmarked so I can complete this later

[Santiago @svpino]

Today, the best closed-source model is Claude 3 Opus (@AnthropicAI).

The best open-source model is Command R+ (@cohere).

It took quite some time to push OpenAI out of the first place, but it already happened.

GPT-5 when?

[NVIDIA AI Developer @NVIDIAAIDev]

NVIDIA and Google DeepMind have collaborated to optimize Gemma models for NVIDIA AI platforms. @GoogleDevs

Experience the power of RecurrentGemma and CodeGemma for domain-specific use cases on the NVIDIA API catalog.

➡️ https://t.co/v7zPyqvosX

[Bindu Reddy @bindureddy]

Now that we have about two dozen LLMs in the market, here are the dimensions that matter when it comes to using them.

Reasoning - Claude 3 Opus beats everything out

Code - GPT-4 is still king here

Cost - Claude Haiku is your best bet

Latency - Claude or a local open-source model is worse best

Fine-Tuning - If you must fine-tune, I would vote for Mistral.

Best Local Model - If your security team throws a fit for no good reason and insists on a local model - Qwen 72B or Smaug-2 (fine-tune on Qwen). Qwen 72B instruct is on top of the human eval leaderboard.

Best small local model - Starling-7b. Again, on top of that leaderboard.

I purposely didn't include the extended context here as it has yet to translate to good context understanding. For now, I prefer sticking to 128K and dealing with it.

[Ethan Mollick @emollick]

Three trends to watch that will shape the future of what AI will mean for us:
1) The unknown capabilities of frontier models
2) Growing evidence of "superhuman" LLM performance in some areas
3) Autonomous agents
Taken together, the implications are large. https://t.co/eSkVjKWmfL

[Ethan Mollick @emollick]

Big AI release day today (though we still haven’t seen a better than GPT-4 class model): updated GPT-4, Gemini 1.5 wide release, now something from Mistral.

Of course what is actually being released remains unclear (seriously, Mistral, mysterious torrents?), but stuff is moving.

[Veloshnee Naicker @VeloshneeN]

Say ‘I do’ to AI! I just read about a new wedding planner tool that splits decisions. No more ‘he said, she said’—just ‘AI said’! Great to see all industries embrace AI into their core functions!
https://t.co/TH8DRSVdi7

[Ethan Mollick @emollick]

The big education crisis caused by AI is not going to be in schools (there was cheating before AI & we can figure out AI uses that boost learning), but after graduation.

White collar work is secretly based on an apprenticeship system that will break

From my book Co-Intelligence

[Veloshnee Naicker @VeloshneeN]

@heyrobinai Great prompts! I need help to be more productive in the mornings; dealing with toddler tantrums doesn't help productivity though! Do you have a prompt to help me deal with that by any chance🤣??

[Fox News AI @FoxNewsAI]

Google Gemini may soon get major changes, including a subscription model and Gmail suggestions on Android https://t.co/KMELjEs0j7

[Zendesk @Zendesk]

🏆 Zendesk was ranked as a leader in Small Business, Mid-Market, and Enterprise for several categories in @G2dotcom's Spring 2024 reports. We were also included in 579 reports and earned 265 badges.

👉 See for yourself why users love us: https://t.co/KDqRLixcjx

[Matthew Berman @MatthewBerman]

I was fortunate enough to interview @JonathanRoss321, the Founder/CEO of @GroqInc. Thank you, Jonathan!

🔥🎥👇

0:38 - Founding Story
3:20 - Groq Chip Memory
6:24 - Chips vs. Cloud
9:28 - Future of AI
11:04 - Where is the Value in AI?
13:45 - Agents & Inference Speed
17:18 - Optimizing Models for Groq
19:32 - Fears and Hope for the Future
22:24 - Bias and Algo Control