tcsenpai

https://github.com/tcsenpai/flint-chart-skill

Here is a skill using that (tested)

But what if our neurobiological reality includes a system that behaves something like an LLM?

It almost seems like we got inspiration from our brain to build neural networks!

I find it amazing, that we can build microprocessors with transistors the size of a DNA molecule

Hello! On Open WebUI using ollama as a backend :)

I guess Ollama needs to update their version, maybe!

Also: I simply asked to the 7B model "hi, who are u":

Thought for 1 seconds Okay, the user is asking "Who are u?" with a casual tone ("u" instead of "you", exclamation mark). Let me break this down.

First, I need to address the question itself. They want to know who I am as an AI assistant. My standard response is that I'm DeepSeek R1, designed to be helpful and handle reasoning step by step. But maybe they're not familiar with AI terms in general. The "u" part might just be their way of speaking, so I should reply politely using "you" form.

Next, considering the context. They started with an HTML tag " "

----

And it hung there without a reply. It might be an OpenWebUI bug but not so promising

I think they should start aiming for 20B models along with 32B and 7B. Usually 7B is enough for a 8GB GPU, 32B requires a 24GB GPU for decent quants (I can fit a 32B with IQ3_XXS but is not ideal) while 20-ish B models (such as magistral or gpt-oss) are a perfect fit for 16GB GPUs

I was very hyped: maybe Google finally did something new, complete, unifying CLI and IDE, a sort of Claude Code Web but as an efficient, IDE-like, local thing.

Then I installed it and it was a VSCode fork.

Looks like Gemma 3 27b is quite creative in fictional scenarios.

https://garden.tcsenpai.com/bookmarks/ai/ai-convos-notes/gem...

Underrated. And magnificent.

I like this. I am imagining a companion extension for chrome/ff that uses you-get as a backend to implement it in a seamless way. Forward thinking idea: imagine going on youtube and have you-get extension bypass the youtube player and playing the content directly without ads. When I say youtube I might also say any other platform.

This is surely useful right now. I wonder what will happens to all the nice X11 tools once Wayland (hopefully soon) will be the golden standard. There are options to enable X11 behaviors in Wayland but I guess that is just a fallback to the insecure implementation.

Update: v 1.1 is out!

- # Changelog

## [1.1] - 2024-03-19

### Added - New `model_tokens.json` file containing token limits for various Ollama models. - Dynamic token limit updating based on selected model in options. - Automatic loading of model-specific token limits from `model_tokens.json`. - Chunking and recursive summary for long pages - Better handling of markdown returns

### Changed - Updated `manifest.json` to include `model_tokens.json` as a web accessible resource. - Modified `options.js` to handle dynamic token limit updates: - Added `loadModelTokens()` function to fetch model token data. - Added `updateTokenLimit()` function to update token limit based on selected model. - Updated `restoreOptions()` function to incorporate dynamic token limit updating. - Added event listener for model selection changes.

### Improved - User experience in options page with automatic token limit updates. - Flexibility in handling different models and their respective token limits.

### Fixed - Potential issues with incorrect token limits for different models.

I applied (for now) a pre-filled table with a 4096 default limit. Users can also specify an upper or lower limit from the UI directly now. Added chunk and recursive summarization too.

Hi! This was a good suggestion! I implemented it in v 1.1 which is already out :)

Speaking of, I made also a youtube summarizer at https://github.com/tcsenpai/youlama

TIL, I am experimenting with PageAssist right now

Personally I use llama3.1:8b or mistral-nemo:latest which have a decent contex window (even if it is less than the commercial ones usually). I am working on a token calculator / division of the content method too but is very early

Not to sound harsh or anything, this is a genuine question: what is Ruby used for nowadays? I am not talking about projects that are using Ruby since a long time but...what is Ruby used for as a go-to tool nowadays?

Can you give more info on this?

The fact that there are other ways to circumvent 2fa highly depends on companies practices. Using fake informations is a good start but even without fake infos I still am trying to regain access to the majority of my 2faed accounts since last December

Well, I guess it depends. In my case, our PR department asked if there was something similar to a devlog, there wasn't so I made it

Each complication over Markdown is utterly interesting to me. Tho I never stick at one due to diffusion

Proton is a blessing. Proton-GE is the natural evolution of the blessing. UMU Launcher is the steroids over Proton-GE.

Nope. I see it is a vertical tab extension: will try it!

I used Arc a lot back when I used MacOS. Was definitely a pleasing experience. I missed it a lot since I switched to Linux/Firefox. I just set up Zen with all my extensions and bookmarks, it behaves very well: let's see if it stands the trial of time, but very nice work. I like the UI

Very nice. I am trying using ollama with it, looks like something is off either on my server or in the way the program calls it: I get the CPU used instead of the GPU (while using ollama in other apps doesn't).

Anyway, impressive

For me the visual aspect is very important. I want to have a fully browseable frame per frame result. Anyway you can avoid using render sync which basically let the simulation free to run as fast as possible with a minimal overhead for the drawing thing each X step

For me the visual aspect is very important. I want to have a fully browseable frame per frame result. Anyway you can avoid using render sync which basically let the simulation free to run as fast as possible with a minimal overhead for the drawing thing each X step

I have a long running Alien simulation since some months (with power on and off based on my pc use).

I started with a "RGB Triad" Initial Setting preset. The incredible thing is that the three types of creatures are forming complex cell-like structures, with different approaches.

I have geometrical "crystals", long strings that try to eat others, spiked structures forming walls...is fascinating.

This is a funny pseudopaper made with Perplexity that me and the AI created based on a very superficial idea I got from a tiktok video.

If the CIA see this, I am available for work. Same goes for other agencies. Otherwise, anyway I am joking and I don't have 500 millions so don't raid my home.