Рубрика: Experiments
-
Inside the Machine: My Journey Reproducing the Scaling Laws for Language Models
After building my dual-RTX 4080 rig (which I covered in my previous post), I felt like a kid with a…
-
Building a Digital Data Scientist: My Local Run with AutoMind
After spending weeks obsessing over scaling laws and raw TFLOPS, I decided it was time to move up the stack.…
-
Beyond the Single Brain: My Attempt to Build a Fabric for Emergent AI Knowledge (ISEK)
Alright, fellow hardware junkies and algorithm enthusiasts. You know my journey: from building my dual-RTX 4080 rig to wrestling with…
-
The Thinking Illusion: Stress-Testing «Reasoning» Models on My Local Rig
We’ve all seen the benchmarks. The new «Reasoning» models (like the o1 series or fine-tuned Llama-3 variants) claim to possess…
-
Beyond the Frame: How I Reproduced SceneCompleter for 3D Scene Generation on My Local Rig
There is a recurring «wall» every AI hobbyist hits when working with Novel View Synthesis (NVS). You generate a beautiful…
-
Mastering the Motion: My Deep Dive into Deformable Neural Radiance Fields (D-NeRF)
One of the most frustrating limits of early Neural Radiance Fields (NeRF) was their «statue-like» nature. They were great for…
-
Beyond Static Knowledge: Implementing RAG Pipelines on My 8TB Local Lab
We’ve all been there: you ask an LLM a question about a recent event or a specific technical paper, and…
-
Fact-Checking the Machine: My Implementation of the ELEVATE Framework
We’ve all seen it: a RAG system retrieves a document, but the LLM still «hallucinates» by misinterpreting a date or…
-
The Death of Cold Starts? Reproducing Contrastive Matrix Completion for Smarter Recs
If you’ve ever opened a new app and been frustrated by its terrible recommendations, you’ve experienced the «Cold Start» problem.…
-
Smarter with Less: My Local Reproduction of Conditional Class Dependencies for Few-Shot AI
One of the most human-like traits is the ability to see a new object once and recognize it forever. Standard…