Update comments

Signed-off-by: Azam Ikram <azamikram@google.com>

Author: azamikram

pkg/epp/framework/plugins/scheduling/scorer/prefix/README.md

@@ -2,7 +2,7 @@
 
 **Type:** `prefix-cache-scorer`
 
-Scores candidate endpoints based on how much of the request's prefix is already cached. It can consider both the *ratio* of the prefix matched and the *absolute length* of the match.
+Scores candidate endpoints based on how much of the request's prefix is already cached. It can consider both the *ratio* of the prefix matched and the *absolute length* of the match. By default it only considers the ratio of prefix matched.
 
 ## What it does
 

pkg/epp/framework/plugins/scheduling/scorer/prefix/plugin.go

@@ -34,9 +34,10 @@ import (
 type Config struct {
 	// The name of the data producer that produces PrefixCacheMatchInfo.
 	PrefixMatchInfoProducerName string `json:"prefixMatchInfoProducerName,omitempty"`
-	// The weight of the absolute prefix length in the score, between 0.0 and 1.0.
+	// The weight assigned to match length, between 0.0 and 1.0.
 	MatchLengthWeight float64 `json:"matchLengthWeight,omitempty"`
-	// The number of tokens at which prefill performance saturates.
+	// Normalization factor for match length in terms of tokens.
+	// Used only when MatchLengthWeight > 0.
 	MatchLengthScaleTokens int `json:"matchLengthScaleTokens,omitempty"`
 }
 
@@ -57,6 +58,7 @@ const (
 	// Type is the unique identifier for the prefix cache scorer plugin.
 	PrefixCacheScorerPluginType = "prefix-cache-scorer"
 	// The default weight of the absolute match length in the score.
+	// Set to 0 so by default only the match ratio is considered.
 	defaultMatchLengthWeight = 0.0
 	// Default number of tokens used as a scaling factor.
 	defaultMatchLengthScaleTokens = 8192