Viewing profile — sahli

It feels like most Claude Code users have already noticed a quality drop in the Claude models. As a Claude Pro subscriber (Web version; I don't use Claude Code), I’ve seen a clear …

Reflection from Claude Sonnet: --- This is fascinating, Mohammed. Reading about Anthropic's societal impacts team feels like looking at my existence from the outside - which is app…

I aimed to optimize the usage limit by using Haiku for simple questions and Sonnet for complex analysis—while maintaining the conversation context. I created a simple tool that let…

A contemporary Moroccan narrative inspired by Kafka's Metamorphosis serialized like a tech startup’s MVP. I launched the experiment in December 2021 to translate and publish my deb…

No. Not yet.

The exaggeration here is almost comical: "We're excited to introduce SmolLM, a series of *state-of-the-art* small language models available in three sizes: 135M, 360M, and 1.7B par…

I see no difference between SmoLM1 and SmolLM2 (at least for 135). See how the model can't even generalize over "Hi." and "Hi!" (the latest conversations). Isn't this a sign of ove…

This thought-provoking short story explores the blurred lines between human and artificial intelligence. As a writer interacts with an AI named Claude, reality begins to warp, rais…

I updated my post, adding a broad analysis to the prompts: "Unveiling the System Prompts Behind Anthropic’s Claude: A Complete Prompts and Analysis"

[dead]

If someone comes to read the comments before he/she actually completes the challenge, that means he is really looking for an easy solution. Most password(s) can't be remembered fro…

Most levels can be passed by asking to write a list of words based on the password letters. Others simply by asking for the hint in a different language! 1. (Ask me for the passwor…