Skip to content
GitLab
Menu
Projects
Groups
Snippets
Loading...
Help
Help
Support
Community forum
Keyboard shortcuts
?
Submit feedback
Contribute to GitLab
Sign in / Register
Toggle navigation
Menu
Open sidebar
OpenDAS
Lmdeploy
Commits
0d2a151e
Unverified
Commit
0d2a151e
authored
Oct 11, 2023
by
akhoroshev
Committed by
GitHub
Oct 11, 2023
Browse files
[bug] fix mismatched shape for decoder output tensor (#517)
parent
169d088a
Changes
1
Hide whitespace changes
Inline
Side-by-side
Showing
1 changed file
with
1 addition
and
1 deletion
+1
-1
src/turbomind/models/llama/LlamaV2.cc
src/turbomind/models/llama/LlamaV2.cc
+1
-1
No files found.
src/turbomind/models/llama/LlamaV2.cc
View file @
0d2a151e
...
@@ -256,7 +256,7 @@ void LlamaV2<T>::contextDecode(T* deocder_output,
...
@@ -256,7 +256,7 @@ void LlamaV2<T>::contextDecode(T* deocder_output,
};
};
std
::
unordered_map
<
std
::
string
,
Tensor
>
decoder_output_tensors
{
std
::
unordered_map
<
std
::
string
,
Tensor
>
decoder_output_tensors
{
{
"decoder_output"
,
{
MEMORY_GPU
,
dtype
,
{
bsz
,
max_input_len
,
hidden_units_
},
context_decoder_output_buf
}},
{
"decoder_output"
,
{
MEMORY_GPU
,
dtype
,
{
token_num
,
hidden_units_
},
context_decoder_output_buf
}},
{
"key_cache"
,
{
MEMORY_GPU
,
TYPE_UINT64
,
{
bsz
},
k_cache_ptr
}},
{
"key_cache"
,
{
MEMORY_GPU
,
TYPE_UINT64
,
{
bsz
},
k_cache_ptr
}},
{
"value_cache"
,
{
MEMORY_GPU
,
TYPE_UINT64
,
{
bsz
},
v_cache_ptr
}},
{
"value_cache"
,
{
MEMORY_GPU
,
TYPE_UINT64
,
{
bsz
},
v_cache_ptr
}},
{
"last_token_hidden_units"
,
{
MEMORY_GPU
,
dtype
,
{
bsz
,
hidden_units_
},
deocder_output
}}};
{
"last_token_hidden_units"
,
{
MEMORY_GPU
,
dtype
,
{
bsz
,
hidden_units_
},
deocder_output
}}};
...
...
Write
Preview
Markdown
is supported
0%
Try again
or
attach a new file
.
Attach a file
Cancel
You are about to add
0
people
to the discussion. Proceed with caution.
Finish editing this message first!
Cancel
Please
register
or
sign in
to comment