diff --git a/src/Init/Data/Float.lean b/src/Init/Data/Float.lean
index f31f76164e..f8153b2f02 100644
--- a/src/Init/Data/Float.lean
+++ b/src/Init/Data/Float.lean
@@ -72,21 +72,35 @@ instance floatDecLt (a b : Float) : Decidable (a < b) := Float.decLt a b
 instance floatDecLe (a b : Float) : Decidable (a ≤ b) := Float.decLe a b
 
 @[extern "lean_float_to_string"] opaque Float.toString : Float → String
-
-/-- If the given float is positive, truncates the value to the nearest positive integer.
-If negative or larger than the maximum value for UInt8, returns 0. -/
+/-- If the given float is non-negative, truncates the value to the nearest non-negative integer.
+If negative or NaN, returns 0.
+If larger than the maximum value for UInt8 (including Inf), returns maximum value of UInt8
+(i.e. UInt8.size - 1).
+-/
 @[extern "lean_float_to_uint8"] opaque Float.toUInt8 : Float → UInt8
-/-- If the given float is positive, truncates the value to the nearest positive integer.
-If negative or larger than the maximum value for UInt16, returns 0. -/
+/-- If the given float is non-negative, truncates the value to the nearest non-negative integer.
+If negative or NaN, returns 0.
+If larger than the maximum value for UInt16 (including Inf), returns maximum value of UInt16
+(i.e. UInt16.size - 1).
+-/
 @[extern "lean_float_to_uint16"] opaque Float.toUInt16 : Float → UInt16
-/-- If the given float is positive, truncates the value to the nearest positive integer.
-If negative or larger than the maximum value for UInt32, returns 0. -/
+/-- If the given float is non-negative, truncates the value to the nearest non-negative integer.
+If negative or NaN, returns 0.
+If larger than the maximum value for UInt32 (including Inf), returns maximum value of UInt32
+(i.e. UInt32.size - 1).
+-/
 @[extern "lean_float_to_uint32"] opaque Float.toUInt32 : Float → UInt32
-/-- If the given float is positive, truncates the value to the nearest positive integer.
-If negative or larger than the maximum value for UInt64, returns 0. -/
+/-- If the given float is non-negative, truncates the value to the nearest non-negative integer.
+If negative or NaN, returns 0.
+If larger than the maximum value for UInt64 (including Inf), returns maximum value of UInt64
+(i.e. UInt64.size - 1).
+-/
 @[extern "lean_float_to_uint64"] opaque Float.toUInt64 : Float → UInt64
-/-- If the given float is positive, truncates the value to the nearest positive integer.
-If negative or larger than the maximum value for USize, returns 0. -/
+/-- If the given float is non-negative, truncates the value to the nearest non-negative integer.
+If negative or NaN, returns 0.
+If larger than the maximum value for USize (including Inf), returns maximum value of USize
+(i.e. USize.size - 1;  Note that this value is platform dependent).
+-/
 @[extern "lean_float_to_usize"] opaque Float.toUSize : Float → USize
 
 @[extern "lean_float_isnan"] opaque Float.isNaN : Float → Bool
diff --git a/tests/compiler/float.lean b/tests/compiler/float.lean
index 882285bc10..60caceed68 100644
--- a/tests/compiler/float.lean
+++ b/tests/compiler/float.lean
@@ -19,12 +19,29 @@ def tst1 : IO Unit := do
   IO.println (0 / 0 : Float).toUInt16
   IO.println (0 / 0 : Float).toUInt32
   IO.println (0 / 0 : Float).toUInt64
+  IO.println (0 / 0 : Float).toUSize
+
   IO.println (-1 : Float).toUInt8
   IO.println (256 : Float).toUInt8
   IO.println (1 / 0 : Float).toUInt8
+
+  IO.println (-1 : Float).toUInt16
+  IO.println (2^16 : Float).toUInt16
+  IO.println (1 / 0 : Float).toUInt16
+
+  IO.println (-1 : Float).toUInt32
+  IO.println (2^32 : Float).toUInt32
+  IO.println (1 / 0 : Float).toUInt32
+
   IO.println (-1 : Float).toUInt64
   IO.println (2^64 : Float).toUInt64
   IO.println (1 / 0 : Float).toUInt64
+
+  let platBits := USize.size.log2.toFloat
+  IO.println (-1 : Float).toUSize
+  IO.println ((2^platBits).toUSize.toNat == (USize.size - 1))
+  IO.println ((1 / 0 : Float).toUSize.toNat == (USize.size - 1))
+
   IO.println (let x : Float := 1.4; (x, x.isNaN, x.isInf, x.isFinite, x.frExp))
   IO.println (let x : Float := 0 / 0; (x, x.isNaN, x.isInf, x.isFinite, x.frExp))
   IO.println (let x : Float := -0 / 0; (x, x.isNaN, x.isInf, x.isFinite, x.frExp))
diff --git a/tests/compiler/float.lean.expected.out b/tests/compiler/float.lean.expected.out
index 02231fc47e..a4f8f92625 100644
--- a/tests/compiler/float.lean.expected.out
+++ b/tests/compiler/float.lean.expected.out
@@ -18,11 +18,21 @@ true
 0
 0
 0
+0
 255
 255
 0
+65535
+65535
+0
+4294967295
+4294967295
+0
 18446744073709551615
 18446744073709551615
+0
+true
+true
 (1.400000, (false, (false, (true, (0.700000, 1)))))
 (NaN, (true, (false, (false, (NaN, 0)))))
 (NaN, (true, (false, (false, (NaN, 0)))))