Discussion
Jack, I mentioned you in my book Unstable Innovation; I hope you can have a quick look at it one day! 🙏
Oh wow, this + Codename Goose has quite the disruptive potential!
Document parsing is a huge unlock
I did test the downloaded LLM and it worked perfectly on a computer that was not connected to the internet. Please be responsible and use the offline uncensored LLM for good rather than evil. Uncensored LLMs can be beneficial if used properly.
Sandisk 128GB Flash Drive
Larger Sandisk SSD
Faraday Cage Box
I am doing some abstract reasoning research via Tiny-Llama on Ollama server, essentially I'm stress testing all the 1.5 billion parameter models
As per research published in the journal Science Advances, TiNyLlamA (Tiny LLama)
is an AI model that excels at abstract reasoning tasks such as finding solutions
to complex problems based on incomplete information. The Tiny LLama architecture
is highly efficient, making it suitable for applications requiring low computation
and memory usage.
Compared to larger models like Google's AlphaZero, TiNyLlamA has a smaller number
of parameters (1.1B) while achieving comparable or better performance on the same
tasks. This makes Tiny LLama an attractive option for industries that require
solutions with limited computational capabilities.
TiNyLlamA has been trained on a variety of real-world problems, including complex
optimization and natural language processing applications. The model's compact
size makes it ideal for deployment on edge devices where memory constraints are
prevalent. TiNyLlamA also has the potential to solve larger datasets with greater
efficiency, making it an attractive option in situations where resources are
limited.
Overall, TiNyLlamA is a promising model with great potential for use in various
applications that require efficient, reliable and accessible machine learning
solutions.
Nice that you can have structured output.
Open Source.
Send it to the lads running DOGE, I think they have a good use case at the moment...
Too complex for them highly regarded individuals
what did you use
This is huge!!