I have ran LLMs locally that had 2048 / 4096 / 8192 context windows That alone keeps me from ever getting one-shot by an AI into a psychosis Being able to tinker with these things is legit good for you CogSec, just saying
@theahmadosman
-
GPU Memory Math for LLMs Explained
By
–
First, GPU Memory Math for LLMs (or why it is not always about Memory Size)
-
Strategic value of open-source AI models and compute costs
By
–
But that’s the thing though, my whole hypothesis was that opensource models are gonna be amazing Waiting until they already are is just reactionary and makes you pay for compute at a steep price
-
Technical Discussion on Model Weighting and User Distribution
By
–
Are you updating the weights for your experiences and focuses vs the normal distribution of all the users a model is being served to
-
The Future of Continual Learning and Local AI
By
–
Continual Learning has already been solved, it just requires the model weights to be running locally on your own hardware so big labs are avoiding the topic altogether Local / Opensource AI will win Inevitable
-
Critique of local AI software stacks and heterogeneous hardware deployment
By
–
Demos in the local AI space between heterogenous devices are misleading -Stuff like mixing MacBooks / Mac Studios with GPUs / DGX Sparks I haven’t seen a single mature software stack in that space yet To talk about it with high certainty is quite dishonest IMHO My 2 cents
-
Critique on the current state of LLM adoption effectiveness
By
–
We still don't know how to use LLMs effectively btw Far from it in fact
-
Preparing for the future of AI Agents and Continual Learning
By
–
In 2024 I tried to tell everyone how to prepare for Agents Last month I tried to tell everyone how to prepare for Continual Learning Get ready
