J'ai tenté brotli, il a l'air de le faire bien, mais en fait il n'est pas capable de le decoder (il esbrouffe), pareil avec zstandard et base64. Par contre il genère les script pour l'encoder et decoder nickel. #chatGPT #encodage
LLMS
-
ChatGPT text compression causes hallucinations warns user
By
–
J'ai tenté la compression de texte avec ChatGPT pour économiser des token, ne faites pas ça, là il hallucine total 🙂 #chatGPT #token
-
LLM Action Generation vs Pre-caching in Sandbox Environments
By
–
Pre-caching input output action is something worth considering. On the other hand, if we let LLM generate actions for any input, we can have an infinite amount of combinations for the outputs, and it seems that it will make the sandbox experience richer.
-
Data Growth and LLM Training: Next Decade Projections
By
–
Sorry I don’t follow. I was respond to you calling LLMs probabilistic, stating we’ve fed it too much data. I think we’ll collect more data in next 10 years than we’ve collected over the last few centuries. Just curious if you’d agree or disagree. Not trying to argue.
-
GPT-4 Text Compression and Decompression Token Efficiency
By
–
yes in many cases the token count is in fact less it is almost impossible to get GPT-4 to regenerate the original text word for word when decompressing the compressed text but usually, the gist of it remains pretty intact
-
Chain of Thought Reasoning and Prompting Techniques
By
–
chain of thought reasoning/prompting https://
arxiv.org/abs/2201.11903 -
Newsletter on prompt engineering techniques and LLM developments
By
–
also, not trying to do an annoying plug, but if you actually want to stay up to date on prompt engineering techniques and stuff I'm working on, check out my newsletter every week I share my analysis on recent developments in LLMs, prompts, and jailbreaks http://
thepromptreport.com -
Jailbreak Techniques for Prompt Engineering and Emergent Behaviors
By
–
these techniques (many initially created for jailbreaks) should be applicable to a range of prompt engineering tasks jailbreaks showcase capabilities on the edge so using similar tactics could reveal other emergent behaviors lmk if there are other techniques I should add!
-
Directional Stimulus Prompting: Using Keywords to Guide LLM Outputs
By
–
Directional Stimulus Prompting we use a smaller LLM to extract keywords from a piece of text and insert them as directional hints into another prompt this is highly beneficial for generating detailed summarizations and I expect more general applications for it will emerge
-
RCI Prompting: Self-Review and Improvement for GPT Outputs
By
–
RCI (Reflect, Critique, Improve) Prompting I haven’t explored this concept too much for jailbreaks but it allows GPT to review its own outputs and fix issues/flaws one area of application this is already immediately helpful is "self-healing" coding and basic reasoning problems