I made an interactive conversation analysis of the (world changing?) #Trump #Zelenskyy #Vance conversation. (Transcript from @apnews.com)

claude.site/artifacts/f8...

Try it yourself — What stands out to you? 👀

The words we choose reveal more than we think. Language isn't just communication — it's a reflection of power, strategy, and worldview.

- Vance emphasized American interests and positioned himself as defending diplomacy.

- Zelensky focused on historical context ("2014," "occupied") and used "we" to suggest collective identity "We are staying in our country, staying strong."

- Trump dominated the conversation (48.1% of total words) and used "you" 28 times, creating a confrontational tone "You don't have the cards. You're buried there."

- Trump's language was heavy on negation and control ("don't," "not," "no").

I analyzed the Trump-Zelensky-Vance Transcript from @apnews.com, to see what stands behind their words - speaking volumes about power dynamics. 📊

Key Insights in thread

OpenAIs o1-pro model is only marginally better than the o1 model. But 10x the price ($20 to $200) to get access, not sure that's worth the qual increase.

Alternatively, do your own agentic or chain of thought prompting. Or go full professional with #dspy or #adalflow for auto prompt optimization.

When predicting human sentiment responses for (german) words, Qwen 2.5 Coder 32B is doing worse than gpt4oMini.

R squared is
0.845 for GPT4oMini
0.714 for Qwen2.4 Coder 32B

Many more comparisons to follow :)

Do you know a good LLM for European Languages?

Using Perplexity instead of prompting to predict the more likely outcome of neuroscience experiments. Beating human experts by roughly 20% - wow.

Could help to make research more efficient Imho. Especially since they've older models to achieve this!

www.nature.com/articles/s41...

Menlo Ventures shows that enterprise usage of anthropic has doubled in the last year.

Could it be because it's often better than openai? - for coding it's my go to model.

A year ago, Claude 2.1 had a 200K token context window with 27% accuracy on needle-in-haystack tests. Now, Alibaba's Qwen2.5-Turbo claims a 1M token window with 100% recall on similar tests.
@alanthompson.net

Performant and actually open LLM was just released. (with training data and recipes etc.) will be testing OLMO soon. allenai.org/olmo

🚀 Just ran a simulation on 1,000 word ratings using human data + our improved LLM prompt for simulating responses. Results? We explain almost 80% of the variance! This approach could revolutionize how we model human emotions—customized for specific groups. 🔥 #AI #EmotionModeling