Accuracy in long document Q&A is important to our customers. Claude 2.1 has demonstrated a 30% reduction in incorrect answers and a 3-4x lower rate of mistakenly concluding a document supports a particular claim.
LLMS
-
Claude 2.1 Doubles Honesty with 2x Fewer False Statements
By
–
Claude 2.1 has made significant gains in honesty, with a 2x decrease in false statements compared to Claude 2.0. This enables enterprises to build high-performing applications that solve business problems with accuracy and reliability.
-
Claude Now Processes 150K Words Entire Documents Analysis
By
–
You can now relay roughly 150K words or over 500 pages of information to Claude. This means you can upload entire codebases, financial statements, or long literary works for Claude to summarize, perform Q&A, forecast trends, compare and contrast multiple documents, and more.
-
Claude 2.1 Launches with 200K Token Context Window
By
–
Our new model Claude 2.1 offers an industry-leading 200K token context window, a 2x decrease in hallucination rates, system prompts, tool use, and updated pricing.
— Anthropic (@AnthropicAI) 21 novembre 2023
Claude 2.1 is available over API in our Console, and is powering our https://t.co/uLbS2JNczH chat experience. pic.twitter.com/T1XdQreluHOur new model Claude 2.1 offers an industry-leading 200K token context window, a 2x decrease in hallucination rates, system prompts, tool use, and updated pricing. Claude 2.1 is available over API in our Console, and is powering our http://
claude.ai chat experience. -
OpenAI News Dominates Weekly AI Briefing Coverage
By
–
This week’s @Digiday AI Briefing was written & edited before all the OpenAI news broke on Friday afternoon, but I still included it as a bullet point on Sunday night before it went live Monday morning. Lots of non-OpenAI news in here too from last week:
-
Switching LLM Providers: Prompting Strategies and LangSmith Hub
By
–
Wondering how to switch LLM providers? It's not as simple as changing the endpoint Different LLMs often require different prompting strategies. One way to explore different prompting strategies? LangSmith Prompt Hub – a collection of different prompts for
-
Inferring LM Prompts from Output Probabilities Without API Access
By
–
haven't posted anything about (2) yet but here's the TLDR: • given LM output probabilities, we can infer what the input prompt was
• we built a model that can do this
• most APIs don't give you probabilities, but we came up with a clever algorithm to get them using logit bias -
Language Model Inversion Research Talk by Sasha
By
–
Sasha gave this amazing talk on our language model inversion research! 1. text embedding inversion (
http://
arxiv.org/abs/2310.06816)
2. language model output inversion (coming soon…) -
Prepare to switch from OpenAI API to other LLMs
By
–
To all AI teams and individuals who have built services on #OpenAI #API, you must be ready at any moment to switch your service to another service. (Anthropic, LLama, any LLM locally on an instance with guardrails in
-
Los Alamos Expands SambaNova AI Deployment for National Security
By
–
COVERAGE: @LosAlamosNatLab expands @SambaNovaAI deployment to run AI workloads for performing national security, science, technology, and engineering projects. Full article via @DanSwinhoe at @dcdnews
: https://
datacenterdynamics.com/en/news/los-al
amos-national-laboratory-expands-sambanova-deployment/
… #ai #generativeai #llm #hpc