Chore: Fix minor doc sentence, grammar, quote errors (#40469)
Signed-off-by: Ashwin Phadke <23502062+ashwin-phadke@users.noreply.github.com> Co-authored-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
Ashwin Phadke committed
Jun 25, 2026 at 00:28 UTC
d511b5bae91423c212457d37936813efeb2646dc
3 files changed
+4
-4
docs/design/hybrid_kv_cache_manager.md
+1
-1
@@ -159,7 +159,7 @@ For simplicity, we assume `block_size=1` in this section.
159
160
### High level idea
161
162
-The block pool uses a dict similar to `tuple(block_hash, group_id) -> block` to catch the full blocks. That means the same tokens of different groups are cached and evicted independently.
162
+The block pool uses a dict similar to `tuple(block_hash, group_id) -> block` to cache the full blocks. That means the same tokens of different groups are cached and evicted independently.
163
164
When a new request comes in, we check the cache hit prefix of each group, and return the intersection of these groups as the cached prefix of the request. See below for the detailed algorithm for checking the cache hit of one group & performing the intersection.
165
docs/design/paged_attention.md
+2
-2
@@ -52,7 +52,7 @@ __device__ void paged_attention_kernel(
52
)
53
```
54
55
-There are also a list of template arguments above the function
55
+There is also a list of template arguments above the function
56
signature that are determined during compilation time. `scalar_t`
57
represents the data type of the query, key, and value data elements,
58
such as FP16. `HEAD_SIZE` indicates the number of elements in each
@@ -178,7 +178,7 @@ const scalar_t* k_ptr = k_cache + physical_block_number * kv_block_stride
178
+ physical_block_offset * x;
179
```
180
181
-Unlike to `q_ptr`, `k_ptr` in each thread will point to different
181
+Unlike `q_ptr`, `k_ptr` in each thread will point to different
182
key token at different iterations. As shown above, that `k_ptr`
183
points to key token data based on `k_cache` at assigned block,
184
assigned head and assigned token.
vllm/v1/engine/__init__.py
+1
-1
@@ -156,7 +156,7 @@ class EngineCoreEventType(enum.IntEnum):
156
class EngineCoreEvent(msgspec.Struct):
157
"""A timestamped engine core event associated with a request.
158
159
- The timestamp is a monotonic timestamps and is used for by the engine
159
+ The timestamp is a monotonic timestamp and is used by the engine
160
frontend to calculate intervals between engine core events. These
161
timestamps should not be compared with timestamps from other processes.
162
"""