Why I love ChatGPT? It's mainly CUSTOM INSTRUCTIONS. Here are the best engineered custom instructions I use everyday:
LLMS
-
Using Motivators in Long Prompts for Better AI Results
By
–
Every longer prompt I write has at least one "motivator". I usually try to pick something that is aligned with the task. I use Adderall when the task is detail-oriented and I want very tight, focused results.
-
LoRAX v0.9 Released with Memory Manager and Qwen 2 Support
By
–
NEW RELEASE: #LoRAX v0.9 is out! This release has the most contributions to date. Thank you to our growing community of contributors!! What's new: Overhauled #LoRA memory manager New model architecture: Qwen 2 Optional max_new_tokens https://
pbase.ai/3VA2nNn -
GPT-5 Launch: Rapid Tutorial Development Expected
By
–
If GPT-5 drops tomorrow, we’ll have tutorials ready by the next day :^)
-
Daily AI Tool Tutorials: Claude 3 and GPT-5 Workflows
By
–
We publish a new worfkflow tutorial every day on the latest tools Right now, we’ve been doing a lot of Claude 3 tutorials (writing style, turning it into an agent, creating animations) But if GPT-5 dropped tomorrow, we’d have tutorials up on it within 24hrs Doing it daily is
-
Potential Bug or Text Retrieval Breakthrough Discovery
By
–
okay what 99.99% chance there's a bug in my code, but
0.01% chance i just solved text retrieval -
Foundation Models and Fine-tuning Explained by Stanford Researcher
By
–
Last year, Ananya Kumar, a machine learning researcher from Stanford University, explained what foundation models are and investigated the common data science practice of fine-tuning these models. https://
youtube.com/watch?v=YzyBSD
n3OQU
… #FoundationModels #DataScience #MachineLearning -
Groq Welcomes Snowflake’s New Leader with Speed Demo
By
–
Hey Sridhar at @SnowflakeDB
, Congrats on the new role! Your predecessor once said everything is better in the #DataCloud. We believe #LLMs and #GenAI run #BetterOnGroq. We think you'll agree so we created a #GroqSpeed demo just for you. https://
hubs.la/Q02qwbmT0 -
Frankenmerge and Continue-Tune Experiment Shows Surprising Success
By
–
Frankenmerge + continue-tune experiment day: surprisingly successful
-
Spend Time Understanding Your AI Model for Your Use Case
By
–
You have to really spend some time with the model and get a feel for it, especially on your use-case