Meet Jamba http://
ai21.com/jamba Build with Jamba https://
huggingface.co/ai21labs/Jamba
-v0.1
… Read more http://
ai21.com/blog/announcin
g-jamba
… #Jamba #Mamba #Transformer #AIarchitecture #SSM
RESEARCH
-
AI21 Labs Introduces Jamba: New AI Architecture Model
By
–
-
Jamba: Revolutionary SSM-Transformer Open Model with 3X Throughput
By
–
Introducing Jamba, our groundbreaking SSM-Transformer open model! As the first production-grade model based on Mamba architecture, Jamba achieves an unprecedented 3X throughput and fits 140K context on a single GPU. Meet Jamba http://
ai21.com/jamba Build on @huggingface -
LPU Speed Applications: Which Use Case Interests You Most?
By
–
We can leverage LPU speed for many use cases. Which are you most interested in?
Multimodal work
General Math
Experimental physics
Molecular chemistry
Computer vision
Combinatorial optimization problems Computational fluid dynamics
Imaging Financial risk calculations
Simulations -
Eigenvalues and Eigenvectors: Principal Component Analysis Explained
By
–
Physical Significance! From Google's PageRank algorithm to PCA they're everywhere! In PCA, eigenvalues quantify data variance captured by each principal component, while eigenvectors define the directions of maximum variance. Implementation below uses eigenvector/values:
-
Eigenvectors and Transformations: Mathematical Foundations for AI
By
–
Now, let's take a real example and put our statement to the test! In the image below, pay close attention to how the transformation affects eigenvectors compared to regular vectors:
-
Eigenvectors and Eigenvalues: Understanding Linear Transformations
By
–
Now, imagine a transformation that changes the space but preserves some directions. These special directions are the eigenvectors, and the scale of stretching or shrinking in these directions is given by eigenvalues. Here's how we calculate them
-
Claude 3 Personality: Intelligence Gain or Training Artifact?
By
–
Claude 3 definitely seems more person-like, but is that a sign that it is (a) smarter or (b) something else? I'm curious if the team at @AnthropicAI may have tuned the RLHF or curated training data to achieve this? Perhaps it's even a result of their focus on safety?
-
Databricks Model Prompt Shows Claude 3 Similarities
By
–
The system prompt for the new Databricks model:
-It seems to have some of the exact same phrasing as the Claude 3 prompt (which they publicly released)
-The prompt tells it to insist that it isn't trained on copyright material, which doesn't actually mean anything but is funny -
Databricks Launches Open Source GPT-3.5 Class Competitor Model
By
–
Databricks launched a open source(ish) GPT-3.5 class competitor. They claim it beats the other open source GPT-3.5 models, and, in quick informal testing, it seems okay and very GPT-3.5. Really, the question is whether more advanced models, coming soon, make this irrelevant.
-
RIP Daniel Kahneman, 1934 – 2024
By
–
RIP Daniel Kahneman, 1934 – 2024 https://
amp.cnn.com/cnn/2024/03/28
/world/daniel-kahneman-nobel-winner-death-intl-hnk
…