Finally, we’ve published a separate high-level note on where we’re hoping mechanistic interpretability research can go. We think it’s good to occasionally step back from our research and reflect on what we're aiming for. https://
transformer-circuits.pub/2023/interpret
ability-dreams/index.html
…
LLMS
-
Mechanistic Interpretability Research Directions and Future Goals
By
–
-
Transformer Circuits Research Updates and Community Comments
By
–
We also list new comments and any corrections on our old papers. We periodically add comments (both from ourselves and others) on previous papers, but they can easily be missed since they're just added to the bottom of existing papers. https://
transformer-circuits.pub/2023/may-updat
e/index.html#new-comments
… -
Replit Launches New Code Complete Model for Ghostwriter
By
–
Why not try out our new code complete model? It's powering Ghostwriter right now, boasting a 40% better performance than the best open-source models. https://t.co/OOn2zSYQAj pic.twitter.com/GV7Uhcjrut
— Replit ⠕ (@Replit) 24 mai 2023Why not try out our new code complete model? It's powering Ghostwriter right now, boasting a 40% better performance than the best open-source models. https://
replit.com/site/ghostwrit
er
… -
LangChain Replicate Integration Simplifies Open Source Model Deployment
By
–
Note that this is not the only option to easily use open source models off the shelf For example we also have a Replicate integration: https://
python.langchain.com/en/latest/modu
les/models/llms/integrations/replicate.html
… In general I'm really excited to see more options like this – makes it much easier to experiment with these OSS models! -
Accessing Open Source Models via API Made Easy
By
–
My biggest hurdle to using open source models is that I just want to access them behind an API to start At some point I may want to fine tune them, but to start I don't want to worry about hosting – I just want to get started building! @MosaicML makes it easy to do so
-
MosaicML Models Integration with LangChain Platform
By
–
First, you do need to sign up for an account. You can do that here: https://
forms.mosaicml.com/demo?utm_sourc
e=langchain
… They have two types of models, both of which are integrated with LangChain: Text Generation Models: https://
python.langchain.com/en/latest/modu
les/models/llms/integrations/mosaicml.html
… Text Embedding Models: https://
python.langchain.com/en/latest/modu
les/models/text_embedding/examples/mosaicml.html
… -
MosaicML Inference API Integration with LangChain Open Source
By
–
Want to be able to easily access open source models behind an API? @MosaicML has got you covered Excited to announce an integration of their inference API with @langchain
, making it easy to experiment with MPT-7B, Dolly, and embedding models Quick -
Replit AI conversation on LLMs with industry experts
By
–
Are you tuning in to our conversation on LLMs with @Chipro @amasad @pirroh today? Join the conversation in this thread, or mention us with #ReplitAI. Let us know all your questions, thoughts, breakthroughs, and takeaways!
-
CommBank leverages 49 LLMs through H2O AI partnership
By
–
“Partnering with @h2oai we've been able to leverage 49 of the best large language models (LLMs) that exist globally today." – Andrew McMullan, Chief Data & Analytics Officer @CommBank http://
ow.ly/VvRF50Ov3QU
#LLMs #GenerativeAI #H2OLLMStudio #CustomerObsessed -
Tree-of-Thought LLMs: Breaking Complex Problems Into Reasoning Paths
By
–
Code for tree-of-thought LLMs. Breaks down complex problems into self-evaluated explorations of different reasoning paths. Has some interesting applications for creative writing. LLMs are starting to develop a sort of internal monologue. https://
github.com/ysymyth/tree-o
f-thought-llm/
…