GPT-4

I'm not sure what is the state-of-the-art in sentiment analysis. However, GPT-3.5 and GPT-4 work great for it with few shot learning. You give a representative set of examples in the prompt and it works pretty well. On the topic of classification, large language models such as BERT are the state-of-the-art. But, there was a paper recently at the ACL [1] which showed that a simple approach of using gzip compression + k-nearest neighbour achieved similar level of performance. Later on, their KNN evaluation methodology came under question because instead of doing KNN with k=2 they reported accuracy based on top-2. Even then, what it shows is that we don't need LLMs for simple tasks such as classification, where traditional techniques still work well and are much cheaper to run. I'm sure some people might say that we should just discard that paper because it had that error. But Sebastian Raschka ran it on the IMDB movie reviews dataset and reported 71% accuracy [3]. I wrote an explainer article on it [2]. [1] https://aclanthology.org/2023.findings-acl.426.pdf [PDF] [2] https://codeconfessions.substack.com/p/decoding-the-acl-pape... [3] https://magazine.sebastianraschka.com/p/large-language-model

I was using mostly Jasper Chat at the end. But when I started comparing Jasper Chat's output to the free version of ChatGPT, I was getting much better outputs with ChatGPT. So, I came to the conclusion that Jasper Chat was still using GPT 3.0 and not GPT 3.5. They never clarified which Jasper Apps were still using which versions of GPT. Therefore, why pay $120 per month (even though I was still on their cherished "unlimited" plan), when they were still obviously giving me inferior outputs with GPT 3.0, when I could get superior outputs with ChatGPT 3.5 for free or pay just $20 per month for ChatGPT 4.0. It's a no brainer

I also notice this with plus account. Over-optimization of inference to reduce costs/increase scale maybe

Qwen3.5 is like an old version of ChatGPT and I can use it the same way I used GPT4 — writing emails, reading documentation and answering questions about it, reviewing code, answering trivia, etc.