51–60 of 62 posts
Earlier quoted context omitted.
It is Llama-2-70b-chat. I quantized it to 2q_k using `quantize` with llama.cpp.
So your experience isn’t representative of the work presented by this post? Or does llama.cpp use the same technique for quantization?
I wonder whether quantization to 1-bit would be functional?
I wonder whether quantization to 1-bit would be functional?
I wonder whether quantization to 1-bit would be functional?
Earlier quoted context omitted.
"For example, the addition of two numbers, 2+2, can be seen as a repeated application of the multiplication operation," I bet most humans ever born did not know what the word "multiplication" meant. Not sure how your metric is useful.
Sometimes, to achieve some perspective, I also zoom out and declare normal over the estimated +-120,000,000,000 people who have ever lived.
Hint: Perhaps read the original output more carefully: "For example, the addition of two numbers, 2+2, can be seen as a repeated application of the multiplication operation,". Or are you counted in the majority of that 120,000,000,000?
Earlier quoted context omitted.
Sometimes, to achieve some perspective, I also zoom out and declare normal over the estimated +-120,000,000,000 people who have ever lived.
I think the point went over your head (which just further proves mine). Hint: Perhaps read the original output more carefully: "For example, the addition of two numbers, 2+2, can be seen as a repeated application of the multiplication operation,". Or are you counted in the majority of that 120,000,000,000?
in the
the spring
Earlier quoted context omitted.
yes. common bit depths to represent pixels range up to 64-bits.
But the physical space?
log2(8,294,400) ≈ 23 (each additional bit doubles the number of unique addresses that can be represented)
so 24 bits are the minimum required to locate a point in 2-d space at the resolution of "4k". when you add in color, 60 bits per pixel.
19 bits for 640x480.