|
--- |
|
license: apache-2.0 |
|
--- |
|
[quants uploading in progress] |
|
|
|
<font color=red><h3> Ultra High Remaster of the incredible: Psyonic-Cetacean-20b. </h3></font> |
|
|
|
This is a Floating Point 32 upscale, where all components and merges were remastered to floating point 32. |
|
This includes all the merges (recreated with master files), and where possible subbing full FP32 models. |
|
|
|
The goal: Carry forward maximum precision right up to the point where it is "GUFFed". |
|
|
|
This includes F32 master file for GGUF too... at a whopping 78 GBs. |
|
|
|
Why? |
|
|
|
Because the difference between F32 vs BF16 is... 8 DECIMAL places. |
|
|
|
And as each merge / model is modified there are "losses" along the way. |
|
|
|
These losses are carried forward and in turn lead to more losses. |
|
|
|
Small? |
|
|
|
Yes... but multipled by each merge(s), and compression(s): 20 billion times. |
|
|
|
<B>The result:</b> |
|
|
|
At Q2K an impressive drop of 533 points in perplexity. (lower is better) |
|
(VS: Q2K orginal base model: PPL = 9.8077 +/- 0.06821 ) |
|
|
|
At Q4KM an incredible drop of 976 points in perplexity. |
|
(VS: Q4km orginal base model -> PPL = 8.7858 +/- 0.06074) |
|
|
|
At Q6 an awesome drop of 234 points in perplexity. |
|
(VS: Q6 orginal base model -> PPL = 8.6070 +/- 0.05907 ) |
|
|
|
To put this in perspective "Q6" now operates ABOVE the orginal full precision version of "Psyonic-Cetacean-20b". |
|
|
|
This because at "Q6" the quant / compressed model is considered to be accurate within "+0.0008 ppl" of the full, uncompressed / unquanted model and it exceeds this by over 200 points. |
|
|
|
<B>The bottom line here is:</b> |
|
|
|
Higher quality instruction following and output. |
|
|
|
Likewise you can use a smaller compression, with higher token per second and still get great quality. |
|
|
|
Same great model... turbo charged. |
|
|
|
This is the first group of remasters. |
|
|
|
<B>More Coming soon...</B> |
|
|
|
It will be followed by a "reg quant plus", which added additional components into the GGUF (all) at floating point 32 |
|
precision to further increases the sheer creatity and raw AI horsepower. |
|
|
|
Following this group will be a full precision Imatrix, and Imatrix Plus repo that will push the limit even more. |
|
|
|
Details of all methods employed to make this high precision remasters will be posted shortly along with comparsions of orginal model and new ultra remaster. |
|
|
