sleepy/llama.cpp
1
0
Fork 0
Code Issues 9 Pull Requests Actions 102 Packages Projects Releases Wiki Activity

docs : Minor cleanups (#19252)

Browse Source 
* Update old URLs to github.com/ggml-org/

* Bump copyrights
This commit is contained in:
Christian Kastner
2026-02-02 07:38:55 +01:00
committed by GitHub
parent b4d05a3d2f
commit 7a4ca3cbd9
36 changed files with 42 additions and 42 deletions
Download Patch File Download Diff File Expand all files Collapse all files
tools/perplexity/README.md
+1 -1
View File
@@ -29,7 +29,7 @@ In addition to the KL divergence the following statistics are calculated with `-
* Mean change in "correct" token probability. Positive values mean the model gets better at prediction, negative values mean it gets worse.
* Pearson correlation coefficient of the "correct" token probabilites between models.
* Percentiles of change in "correct" token probability. Positive values mean the model gets better at prediction, negative values mean it gets worse. Can be used to judge noise vs. quality loss from quantization. If the percentiles are symmetric then the quantization is essentially just adding noise. If the negative values are significantly larger than the positive values then this indicates that the model is actually becoming worse from the quantization.
* The root mean square of the change in token probabilities. If you were to assume that the quantization simply causes Gaussian noise on the token probabilities then this would be the standard deviation of said noise. The uncertainty on the value is calculated that the change in token probabilities follows a Gaussian distribution. Related discussion: https://github.com/ggerganov/llama.cpp/discussions/2875 .
* The root mean square of the change in token probabilities. If you were to assume that the quantization simply causes Gaussian noise on the token probabilities then this would be the standard deviation of said noise. The uncertainty on the value is calculated that the change in token probabilities follows a Gaussian distribution. Related discussion: https://github.com/ggml-org/llama.cpp/discussions/2875 .
* Same top p: Percentage of how often the token was assigned the highest probabilites by both models. The uncertainty is calculated from the Gaussian approximation of the binomial distribution.
## LLaMA 3 8b Scoreboard