Sam Altman posted that GPT-4's performance should be ramped up after its reported 'laziness' during the holidays. Even AI struggles with New Year's resolutions:
LLMS
-
Thesaurus Alternative to ChatGPT: Explainability and Scaling Laws
By
–
Perhaps you should publish this result? The prospect of replacing ChatGPT with a thesaurus is enticing, especially for explainability. What do scaling laws for thesauri look like?
-
Predibase AI and LLM Predictions for 2024
By
–
ICYMI: Here are our AI and #LLM predictions for 2024! Sneak Peek: small is the new big, #opensource is the future, synthetic data for everyone, #frankenmodels, and more!
-
Coze AI Platform Features and Workflow Capabilities
By
–
Key features of Coze include a visual builder for bot workflows, allowing multiple bots to interact within a single chat environment. Similar to how the Mentions feature works on ChatGPT but within the context of a single bot
-
Granular control and model selection in Coze AI agents
By
–
Advanced settings in Coze offer granular control over bot responses, with options for temperature adjustment and variable use for logic. GPT4 and GPT-4-turbo are also available. For FREE.
-
ByteDance Launches Coze AI Bot Platform
By
–
It is not viral on AI Twitter but it should be ByteDance (a company behind @tiktok_us
) has launched Coze, an AI bot platform that directly competes with OpenAI's GPTs, enabling users to create and deploy custom AI bots. For FREE. -
How Can LLMs Handle Thesaurus Translation Tasks?
By
–
You can ask LLMs to translate from Lisp to Python, or to invert a string. How would this work with a thesaurus?
-
Understanding as Mapping: LLMs and Causal Model Weights
By
–
I think of understanding as a mapping between observables and causal models. Weights can be a way to represent understanding. LLMs don't always understand, but Gary is himself simulating understanding via prompt completion, which is why he always gets to the same conclusion.
-
Training tiny character RNN for Office script generation
By
–
update: training a tiny char RNN to write lines from The Office (no BOS token yet so these samples always start with M; using ancestral sampling for now)
-
LangChain Deep Dive: LLMs, Agents, and Evaluations with Lance Martin
By
–
GenAI solutions with LangChain: Lance Martin on LLMs, agents, evals, and more! Thanks to @neo4j for having our very own @RLanceMartin on their http://
GraphStuff.FM podcast! A great deep and technical dive into a lot of advanced topics! https://
graphstuff.fm/episodes/genai
-solutions-with-langchain-lance-martin-on-llms-agents-evals-and-more
…