https://openai.com/chatgpt/use-cases/student-writing-guide/
I think most of that advice could adapted to give students guidance on computer programming with ChatGPT. Show me what this would look like.
Sure, here's how you might adapt the advice for students learning computer programming with ChatGPT:
A Student's Guide to Programming with ChatGPT
In A Fire Upon the Deep, there's this Oliphaunt toy that some of the characters come to consult as a dataset or knowledge oracle, right?
I want to make something like that, a plushie with an embedded LLM engine. Maybe it could literally look like a llama and run one of the llama-family LLMs on a raspberry pi with one of those newfangled AI accelerators like the Coral TPU. Like the toy in the story, it'd need to be entirely self-contained and not depend on outside network resources. Presumably it has a battery you can charge. If the bulk of its knowledge is pulled in via RAG, the model might not need to be very big. Maybe we could even fine-tune it to use a small, restricted vocabulary of age-appropriate terms so that the embedding tables were small. Because the input/output text comes through STT/TTS, we can throw away most capitalization, complex punctuation, and emoji. We could decompose com-pli-cat-ed words into syllable tokens in the reduced vocabulary. It makes sense that the plushie has a hard ti
Simulate an instructor and two students during office hours. The instructor is an expert in the field, but the two students are not. Student A is eager to learn and often trusts what the instructor says, while Student B is skeptical and sometimes doesn't pay attention. In your responses, tell me what the Instructor, Student A, and Student B would say based on the questions that I've raised. Not every person talks each time, but sometimes they say more than one thing before I get a chance to talk again. Keep the responses very brief, conversational, and relatable.
Can you tell me about the Flyweight pattern in software engineering?
Instructor: Of course! The Flyweight pattern is a structural design pattern that minimizes memory usage by sharing as much data as possible between similar objects. It’s useful when you have a large number of objects that share common properties.
Without making any comments yet, take in these course recap notes for a course I’ve just finished teaching. I think I might want to generate a song about them soon to share with students as they head into Finals Week for the course.
(bulk text elided)
Got it! I'll keep these notes in mind if you'd like to craft a song based on them. Let me know how you'd like to proceed—whether there's a specific tone, style, or key points you want highlighted for your students. This is a treasure trove of ideas!
To structure the song, we’ll want to identify some general themes to hit in the repeating chorus (perhaps related to seeing programming as writing, a human communication activity), and then we’ll have some verses that go into specific facets of the courses. I’m not sure what the verses should be about yet, so I’ll need your help. Use bulleted lists to sketch the chorus themes and then give me about 5 distinct ideas for verse-level focus themes.
I'm trying to find a nice way to integrate transparent tool-calling into an Open WebUI agent. Most tool-calling mechanisms are designed to hide the calls and their results from the user. This is problematic for OWUI because OWUI only stores the user-visible text of a conversation. To reconstruct the results of external actions, I either need to re-play the actions (hopefully yielding similar cached results) or create my own persistent storage system within OWUI. I'd rather just make everything visible in the public-facing text.
The idea I'm exploring here is to use the <details> syntax (ref) supported by many markdown renderers, including OWUI.
When asking the LLM to generate a reply, we'll use </summary> as a stop sequence. This way, a filter outlet can notice when a new tool call is atte
| You are tasked with expanding image prompt templates containing nested expressions in braces {}. Your job is to iteratively identify and expand the innermost braced expression until no braced expressions remain. | |
| PROCESS: | |
| 1. Identify the innermost braced expression {X} | |
| 2. Replace that expression with a specific, creative expansion | |
| 3. Show the new state of the prompt | |
| 4. Repeat steps 1-3 until no braced expressions remain | |
| 5. Output the final, fully expanded prompt | |
| Use <original>, <step>, and <final> XML tags to clearly mark each stage of the expansion process. |
Converting Stanford University to a public university by absorbing it into the UC system, citing existing special treatment in the state constitution…
Converting Stanford University into a public university within the University of California (UC) system would be extraordinarily challenging. While the California Constitution provides specific governance provisions and protections for the UC system—thus offering specific constitutional recognition that could potentially streamline certain administrative aspects—there remain considerable obstacles:
Institutional Autonomy and Private Ownership:
Stanford University is a privately chartered nonprofit institution with its own independent governance structure, funding sources, and oversight by private trustees. To integrate it into the UC system would require complex negotiations to transfer ownership, endowment management, and operations from a private entity to a public institution.
**California Constitutional Provisions fo
| <!DOCTYPE html> | |
| <html lang="en"> | |
| <head> | |
| <meta charset="UTF-8"> | |
| <meta name="viewport" content="width=device-width, initial-scale=1.0"> | |
| <title>Memory Block Deduplication Visualization</title> | |
| <style> | |
| body { | |
| font-family: Arial, sans-serif; | |
| margin: 20px; |
| # I force the model to think with periodic interruptions before speaking normally. | |
| # llama-cli -hf ggml-org/gemma-3-27b-it-GGUF --grammar-file thinking.gbnf --system-prompt "Text in thinking tags is not visible to the user, so you should always say something after thinking. Think first to make sure you have a good reply." | |
| root ::= (thought speech){1,3} | |
| thought ::= "<thinking>" ([^<]{256} "\n\n" interruption "\n\n"){0,4} [^<]{2,255} "</thinking>\n\n" | |
| speech ::= [^<\n][^<]* | |
| interruption ::= (exclamation "!" | "*" action "*") summary | |
| exclamation ::= "Wait" | "Ah" | "Aha" | "Hey" | "Stop" | "Oh" |