With projects like babyAGI and Auto-GPT, what other software pieces need to be built to actually achieve effective communication within different LLM’s? @tanishabassan
LLMS
-
Training LLMs with Code Execution Traces Improves Reasoning
By
–
People say that adding code to the LLM training corpus improves the model's "reasoning" ability. If so, would including a training set of short programs followed by an execution trace improve its reasoning further as it would have to predict the executed code path from the src?
-
Small Organizations Training Custom LLMs: Infrastructure Requirements
By
–
How difficult is it for small organizations to tune/build their own LLMs? Will every company eventually have its own little datacenter to train these LLMs? @husseinmuhaisen @GabiImmelman
-
LLM Panel Revival: Answering Audience Questions
By
–
We had such a great conversation about LLMs this week, but we weren’t able to get to all of our audience questions. We’re reviving the panel to let @pirroh @chipro @amasad answer some of them!
-
Programmatic Data Development for Foundation Model LLMs
By
–
Are you looking to power #LLMs with your data? Join us at https://
future.snorkel.ai where @ajratner will discuss how programmatic data development approaches are used for foundation model development. -
GPT-4 Word Length Accuracy Beyond Traditional Counting
By
–
right, GPT-4 might not "count" the letters in a word in the traditional sense, but it's highly accurate when you zero shot ask it for the length of a certain word there's def more to the problem than just GPT's supposed inability to determine the length of a word
-
Databricks Dolly Model Serving Integration Now Available
By
–
Ever since @databricks announced Dolly I was hoping to have an integration with their model serving endpoint! Thanks to Xiangrui Meng and Gengliang Wang!
-
Ghostwriter Code Model Now Supports 20 Programming Languages
By
–
Our new code complete model has been trained on 20 different programming languages, so when you're using Ghostwriter to 10x your workflow you can now do it with confidence in a bigger array of languages.
— Replit ⠕ (@Replit) 26 mai 2023
Try it out right now on Replit. https://t.co/OOn2zSYQAj pic.twitter.com/SS6oWN3l34Our new code complete model has been trained on 20 different programming languages, so when you're using Ghostwriter to 10x your workflow you can now do it with confidence in a bigger array of languages. Try it out right now on Replit. https://
replit.com/site/ghostwrit
er
… -
Falcon 40B Dethrones LLaMa on Open Leaderboard
By
–
LLaMa is dethroned A brand new LLM is topping the Open Leaderboard: Falcon 40B *interesting* specs:
– tuned for efficient inference
– licence similar to Unity allowing commercial use – strong performances
– high-quality dataset also released Check the authors' thread https://
x.com/slippylolo/sta
/slippylolo/status/1662082035744227330
… -
State of GPT: Karpathy Explains LLM Training Paradigms at Microsoft Build
By
–
State of GPT – Andrej Karpathy at Microsoft Build Such awesome talk on state of GPT and other LLMs. Karpathy in his usual best teaching style talks about the training paradigms of large language models(pretraining, supervised finetuning, reward modelling, RLHF), prompting