"OpenAI said before that evaluating their dataset was one of the major factors for the huge jump from ChatGPT3.5 to 4. Let's do the same for our own applications!" @SimonNom1 expressed better than I could why we're focusing so much on end-to-end evaluations!
TOOLS
-
Amazon Bedrock LLMs Integration Represents Major Cloud AI Development
By
–
Amazon bedrocks LLMs are a big integration!
-
Custom Agent with Python REPL and Retriever Tools
By
–
Then we could go about improving it! Our final solution was a custom agent with two tools: – Python REPL (for doing analysis)
– Retriever (for dealing with text data present) We're excited to see other's solutions for this as well! -
LangChain Benchmarking Question-Answering CSV Data Tasks
By
–
Recap of important links: Blog: https://
blog.langchain.dev/benchmarking-q
uestion-answering-over-csv-data/
… YouTube: https://
youtube.com/watch?v=jGnf4O
hptbA
… Code & data used: https://
github.com/langchain-ai/l
angchain-benchmarks
… We had a lot of fun doing this and learned a lot – we're going to do it for more tasks! Up next: SQL -
Streaming Output with cURL and jq Programming Example
By
–
You can also receive streaming output using the programming language of your choice. Here's an example using cURL and jq: pic.twitter.com/6pLpTULCU9
— Replicate (@replicate) 14 août 2023You can also receive streaming output using the programming language of your choice. Here's an example using cURL and jq:
-
API now supports server-sent event streams for language models
By
–
Our API now supports server-sent event streams for language models, giving you live output as the model is running. Learn how to consume event streams to make your apps more responsive.
-
Claude’s 100K Token Context and File Upload Features
By
–
Claude’s 100K token context window and file upload feature unlock many possibilities.
— Anthropic (@AnthropicAI) 14 août 2023
Choose a long-form document like a lease, then drag-and-drop it into Claude for a fast and friendly summary!
Try it today in the US and UK at https://t.co/RxKnLNNcNR. pic.twitter.com/dDiLi3rIw3Claude’s 100K token context window and file upload feature unlock many possibilities. Choose a long-form document like a lease, then drag-and-drop it into Claude for a fast and friendly summary! Try it today in the US and UK at http://
Claude.ai. -
Undo Redo Actions Now Persist Between Repl Reloads
By
–
7. Undo and redo actions now persist between reloads. If you ever need to reload your Repl, you’ll find that picking up and continuing with your work is now instant as your CMD + Z muscle memory will be preserved. pic.twitter.com/K1c5izoQsh
— Replit ⠕ (@Replit) 14 août 20237. Undo and redo actions now persist between reloads. If you ever need to reload your Repl, you’ll find that picking up and continuing with your work is now instant as your CMD + Z muscle memory will be preserved.
-
Faster Deployment for Repls with Large Package Numbers
By
–
4. We've made deployments even better by cutting down the total time to deploy. Repls with a large number of packages deploy 2-3x faster. We’ve also squashed bugs on failed deployments and upgraded the UI and experience. pic.twitter.com/RHqC1ZOBQY
— Replit ⠕ (@Replit) 14 août 20234. We've made deployments even better by cutting down the total time to deploy. Repls with a large number of packages deploy 2-3x faster. We’ve also squashed bugs on failed deployments and upgraded the UI and experience.
-
Repls Launches 256GB Expandable Storage for ML Model Training
By
–
1. First, Expandable Storage: Repls can now store up to 256GB. Pick a plan, or buy storage à la carte. Train your ML models and don’t worry about the size of your data. pic.twitter.com/vpMyKgJ3vH
— Replit ⠕ (@Replit) 14 août 20231. First, Expandable Storage: Repls can now store up to 256GB. Pick a plan, or buy storage à la carte. Train your ML models and don’t worry about the size of your data.