LLMS

→ View original post on X — @replicate

Sub-second cold-boot speeds for SDXL and Llama 2 models

By

@replicate

–

20 October 2023 18h47

In case you didn't know,
Our cold-boot speed is getting low,
For fine-tunes, SDXL and Llama 2
Under one second, they boot for you.

→ View original post on X — @fofrai

Model Distillation Safety Concerns in AI Development

By

@fofrai

–

20 October 2023 18h13

Are there any other models being distilled? The Dreamshaper one is, uh, very unsafe for work.

→ View original post on X — @miketamir

Microsoft PromptFlow: Build Production-Ready LLM Applications

By

@miketamir

–

20 October 2023 17h37

microsoft/promptflow: Build high-quality LLM apps – from prototyping, testing to production deployment and monitoring. https://
bit.ly/46iENqu #AI #MachineLearning #DeepLearning #LLMs #DataScience

→ View original post on X — @groqinc

Groq’s LLM Inference Hardware 10X Faster Than Nvidia

By

@groqinc

–

20 October 2023 17h26

"
@csTimSears dazzled us with a deep dive into @GroqInc
's cutting-edge SW/HW ecosystem. He showcased the lowest latency version of LLM inference hardware, which boasted a speed 10X faster than NVidia. The room was abuzz with questions and excitement, and rightly so."

→ View original post on X — @cohere

Building Chatbots with Chat RAG and Memory

By

@cohere

–

20 October 2023 17h15

We’re excited to host our next Maker Spotlight live demo session today! Join Cohere community champions Arjun Patel and @Coffee_and_NLP in conversation with AI engineer @an1z8 as he walks us through leveraging Chat + RAG to build chatbot applications with memory and context.

Zero-shot Chain of Thought and Prompt Engineering Techniques

By

–

20 October 2023 14h14

Some other prominent techniques include zero-shot chain of thought (beneficial for math or logic-related queries) and taking a step-by-step approach. Do not forget that a lot of this involves trial and error, continued practice, and patience.

Optimize LLMs: Test Models, Prompts, and Parameters Strategically

By

–

20 October 2023 14h14

Maximize LLMs by actively trying different models, prompts, and parameters. Don't worry if it's daunting. Over time, you'll discern which tasks best match each LLM and its optimal settings.

Prompt Engineering Strategies to Boost LLM Performance

By

–

20 October 2023 14h14

Prompt engineering enhances LLM efficiency, adjusting creativity, simplicity, and role clarity. However, strategies effective for one model may not suit another. Few-shot and chain-of-density prompting can boost LLM performance (learn more in the video).

Tokenizers Impact on Model Language Understanding

By

–

20 October 2023 14h14

Note that different tokenizers (the step where you transform words into understandable numbers for the model) influence their sentence comprehension, too!