In his talk at Lllamapalooza NYC, @dawoodnyc
, co-founder of @Gradio (Acq. by @huggingface
) shared insights into building reliable applications with large language models (LLMs), specifically focusing on the Gradio Playground, powered by Cerebras Inference Check out the
CODE
-
Building Reliable LLM Applications with Gradio Playground
By
–
-
Berkeley Research on AI Evaluations and Agent Systems
By
–
Two of the things we spend the most time thinking about are evals and agents This new research from @sh_reya and Ankush Garg at Berkeley touches on BOTH those things Includes an example notebook for how to run on your LangGraph app!
-
Barbara Liskov Birthday: MIT Professor and Turing Award Winner
By
–
Happy bday to MIT Prof. Barbara Liskov, Turing winner & programming pioneer: https://t.co/N1oK03hrUQ pic.twitter.com/qNalDmqeIs
— MIT CSAIL (@MIT_CSAIL) 7 novembre 2024Happy bday to MIT Prof. Barbara Liskov, Turing winner & programming pioneer: http://
bit.ly/2froAuJ -
HoloViz 0.17.3: High-Level Python Visualization Tools
By
–
High-level tools to simplify visualization in Python — HoloViz 0.17.3 documentation https://
bit.ly/3VXvmJC
#AI #MachineLearning #DeepLearning #LLMs #DataScience -
GPU Kernel Memory Optimization Challenges in AI Development
By
–
I know what you mean. For me there was nothing between 1x and 70x, it either worked or it didn't — had to redesign it each time. Now running out of shared memory for the GPU kernel, just a tad too small to be useful!
-
Triton Library Lacks 3.0-Level Maturity: API Limitations
By
–
I don't just mean the limitations of the GPU, that's part of it but done that before. Triton itself doesn't feel like a 3.0-level library. Stuff like http://
tl.dot() doesn't work when you specify the output array as a parameter (per docs), you have to use the retval. -
Triton MLP Implementation Achieves 70x Performance Improvement
By
–
Triton is exciting to work with. I built a type of MLP with it and it's 70x faster than the Torch version… Conversely, the set of features that actually works is so razor thin you have to code-golf every single decision for hours! https://
github.com/triton-lang/tr
iton
… -
ReMoE Implementation Optimization Challenges Wall-Clock Performance
By
–
Thought the same thing and was trying to reimplement ReMoE over the past days… Achieving the same level of utilization as the default codebase, and thus beating the wall-clock record, will not be trivial.
-
Open source project reaches 2K: Animation creation techniques
By
–
我一直以为 10K+了，没想到刚到 2K ，开源不易啊，果断支持一波。 另外，动图怎么弄的
-
Eric Schmidt predicts AGI with 80-90% expertise in 5 years
By
–
Dr. Eric Schmidt, ex-Google CEO says in an interview that within the next 5 years we will have an AI entity that is in a sense AGI and will have 80-90% of all (!) expertise of all disciplines!
— Chubby♨️ (@kimmonismus) 7 novembre 2024
It will also write its own code!pic.twitter.com/N6bu82uh1ZDr. Eric Schmidt, ex-Google CEO says in an interview that within the next 5 years we will have an AI entity that is in a sense AGI and will have 80-90% of all (!) expertise of all disciplines! It will also write its own code!