If what’s keeping you from AI is the impenetrable math, this wonderful book by @anilananth is the antidote.
CODE
-
Grimoire: An AI-Powered Coding Assistant Tool
By
–
5. Grimoire Need a partner for writing code or bug solving assistant? Try this one. Link: https://
chatgpt.com/g/g-n7Rs0IK86-
grimoire
… -
Scaling Engineering Teams: Why More Bodies Don’t Mean Faster Code
By
–
maybe more relatable…. 1 engineer can make a simple code change in 10mins, but 10 engineers cannot change the same part of the code in 1 min. 1 engineer can change a light bulb in 10 mins, but 10 engineers cannot change the same light bulb in 1 min.
-
Indexify: New AI Data Processing and Indexing Platform
By
–
Announcing Indexify https://
bit.ly/3zD4szc
#AI #MachineLearning #DeepLearning #LLMs #DataScience -
Training stability control with loss and gradient norm thresholding
By
–
Cool! For the spike I'd try e.g. `-sl 7 -sg 7` to keep instability in check earlier in the training. (will skip update if loss/gradnorm > 7 sigma outlier is detected)
-
Train Small Language Models Efficiently with Synthetic Data
By
–
Want to train a #small language model for your use case, but short on data and training #infra? Good news: Predibase + @gretel_ai make it easy to create high-quality #synth data and #finetune + serve #SLMs, cost efficiently. Check out our tutorial: https://
pbase.ai/3WnfEc7 -
LLM Capabilities and Abacus Solutions for Enterprise Use
By
–
LLMs are incredibly powerful tools. Here is a small snippet of what they can accomplish today: •Respond to customer service requests
•Write proficient SQL
•Analyze documents
•Write and code and talk to systems
•Generate legal documents At Abacus, we can provide LLMs that -
Layer Replication and Pruning Techniques for Model Optimization
By
–
These experiments re-confirm a lot of recipes discovered by doing frankenmerges. – Replicating layers is helpful for creative tasks but not for reasoning (in general) – Changing the layer order doesn't work – Pruning works best with middle layers but still requires healing
-
Instruction Data Generation: Combining AgentInstruct and Wizard Learning
By
–
I wrote an article on @huggingface about instruction data generation. It analyzes how AgentInstruct and Wizard Learning work, and then proposes a simple architecture to combine their respective strengths. Article: https://
huggingface.co/blog/mlabonne/
agentic-datagen
… -
Adding Memory to Apps with Clarifai Vector Store
By
–
Hi, Adding memory to your app and constantly updating it with your previous conversations will solve this. You can leverage the Clarifai Apps for this, as apps acts as your vector store and you can keep updating the chat to the Clarifai app and get the relevant info based on the